Visual detection device and equipment for live-line state of power transformation equipment

Abstract

The invention discloses a power transformation equipment live-line state visual detection device and equipment, and relates to the field of power detection. The problem that limitation exists in a traditional operation and maintenance mode, especially manual contact type detection is solved. The device comprises a power supply module, an information acquisition unit, a data processing unit and a data visualization unit, the information acquisition unit comprises a visible light camera, a laser radar and an electric field sensor; the data processing unit receives data of the visible light camera through a USB interface, receives data of the laser radar through the Ethernet, and receives data of the electric field sensor through a Modbus RTU communication protocol. The data visualization unit is connected with the data processing unit; and the power supply module is used for providing electric energy for the information acquisition unit, the data processing unit and the data visualization unit. The method is applied to the field of substation equipment maintenance.

Description

Technical Field

[0001] The invention relates to the field of electric power detection, and in particular to a visual detection device for the live state of a transformer equipment. Background Art

[0002] In the operation of the power system, substation equipment plays a vital role, which is directly related to the safe, stable and efficient operation of the power system. Substation equipment is not only a key link in power transmission and distribution, but also involves various aspects of the power system, such as current regulation, load distribution, fault protection, etc. With the rapid development of social economy, especially driven by industrialization and informatization, the demand for electricity in society continues to grow, and the power system is facing unprecedented pressure. In order to meet the growing demand for electricity, substation equipment has to bear more and more loads. However, substation equipment is expensive and generally has a long service life. Therefore, many power companies do not frequently replace or update equipment.

[0003] Due to funding and technical limitations, many power companies, especially small and medium-sized enterprises, are unable to update outdated or aging substation equipment in a timely manner. This situation has led to severe challenges to the operating environment and service life of the equipment. Substation equipment is prone to aging, wear and tear when it is running at high load for a long time, which leads to equipment performance degradation and even failure. This not only affects the speed and quality of power transmission, but may also trigger a series of chain reactions, such as power outages, equipment damage, and even serious threats to the lives of workers. Some old equipment may overheat, short-circuit or electrical failure under long-term overload operation, resulting in serious accidents such as power outages and tripping.

[0004] In this context, the daily operation and maintenance of substations is particularly important. In order to ensure the long-term stable operation of substation equipment, operation and maintenance personnel need to inspect, maintain and repair key equipment such as busbars, switches, and switches. However, traditional operation and maintenance methods still rely on manual operation, especially close-range contact detection methods, which not only requires staff to have superb technical and operational capabilities, but also faces great safety hazards. Since substation equipment is often located in live areas, operation and maintenance personnel are prone to mistakenly enter high-voltage areas during inspections, especially when conducting contact live detection, there is a serious risk of electric shock.

[0005] In recent years, electric shock accidents caused by improper operation or negligence of operation and maintenance personnel have occurred frequently, causing huge casualties and economic losses to power companies. Relevant statistics show that the economic losses caused by power outages, equipment damage and subsequent maintenance caused by substation equipment operation and maintenance accidents have even reached hundreds of billions of yuan. These accidents usually not only damage the equipment itself, but also cause large-scale power outages, affecting the normal operation of social production and life, and bringing broader social and economic impacts.

[0006] With the continuous expansion of the scale of the power grid and the gradual advancement of the reform of the power system, the traditional operation and maintenance methods are becoming less and less suitable for the needs of modern power systems. In the past, the operation and maintenance mode, which mainly relied on manual inspection and contact detection, has been unable to meet the requirements of the increasingly complex, large-scale and intelligent power system. The high-altitude working environment, long-term manual operation and heavy labor intensity have greatly increased the work safety risks of operation and maintenance personnel. Especially in some remote areas or places with complex geographical environments, there are often blind spots in the inspection of substation equipment, and manual inspections cannot cover all areas, resulting in an increased risk of missed inspections or wrong inspections.

[0007] At the same time, the traditional manual operation and maintenance model also faces the problems of high-intensity labor and occupational health of operation and maintenance personnel. Since substation equipment is sometimes installed at high or difficult-to-reach locations, operation and maintenance personnel need to climb power poles and use hanging baskets to conduct inspections. This high-altitude operation not only increases the risk of operations, but also makes workers prone to fatigue and stress. If they are in this working environment for a long time, accidents are likely to occur, and even cause health problems for operation and maintenance personnel. In addition, some substation equipment such as busbars are usually located in complex geographical environments, and operation and maintenance personnel are often difficult to reach. This makes it impossible for manual operation and maintenance to cover all equipment, resulting in problems such as missed detection and false detection, affecting the safe operation of equipment.

[0008] Therefore, the operation and maintenance of substation equipment is not only a technical inspection of the equipment, but also needs to take into account the safety, health and work efficiency of the operation and maintenance personnel. Traditional operation and maintenance methods, especially manual contact detection, have gradually exposed their limitations, and there is an urgent need to seek more efficient, safe and intelligent solutions. Summary of the invention

[0009] In view of the limitation of traditional operation and maintenance methods, especially manual contact detection, the present invention proposes a visual detection device for the live state of substation equipment, which includes:

[0010] Power module, information acquisition unit, data processing unit and data visualization unit;

[0011] The information collection unit includes: a visible light camera, a laser radar, and an electric field sensor;

[0012] The data processing unit receives data from the visible light camera through a USB interface, receives data from the laser radar through Ethernet, and receives data from the electric field sensor through a Modbus RTU communication protocol;

[0013] The data visualization unit is connected to the data processing unit;

[0014] The power supply module is used to provide electric energy to the information acquisition unit, the data processing unit and the data visualization unit.

[0015] Furthermore, a preferred method is proposed, wherein the visible light camera is used to collect image data of the substation equipment and transmit the information to a data processing unit; the lidar is used to collect three-dimensional point cloud data of the substation equipment and transmit the point cloud data to the data processing unit.

[0016] Furthermore, a preferred embodiment is proposed, wherein the data processing unit is an embedded development board, and a YOLO target detection algorithm is embedded in the embedded development board.

[0017] Furthermore, a preferred embodiment is proposed, in which the processing flow of the data processing unit is as follows:

[0018] The embedded development board receives the images collected by the visible light camera, the point cloud data acquired by the lidar, and the field strength data returned by the electric field sensor;

[0019] Perform target detection on the collected images, identify the substation equipment in the images, obtain the category information of the substation equipment, mark the substation equipment with a detection box, and calculate the pixel values ​​of the four vertices of the detection box;

[0020] Filter the collected point cloud data;

[0021] The point cloud data inside the detection frame is classified using Euclidean clustering, and the center point is fitted as the center point of the substation through the external bounding box method, and the distance of the substation is obtained through the center point;

[0022] The distance between a person and the substation is calculated using the electric field strength measured by the electric field sensor and the voltage of the substation, and compared with the distance to the substation obtained through the central point to determine the specific location and energized state of the energized equipment.

[0023] Furthermore, a preferred method is proposed, wherein the electric field sensor is used to continuously measure the AC power frequency electric field near the 110kV AC voltage line to obtain the electric field strength information radiated outward by the substation equipment.

[0024] Furthermore, a preferred method is proposed, in which the electric field detection range of the electric field sensor is -50kV / m to 50kV / m, which is used to obtain the electric field strength of the substation equipment that radiates electric field outward from the surrounding area, wherein the embedded development board is equivalent to the host in the modbus communication mechanism, and the electric field sensor is equivalent to the slave. The host sends a read instruction to the slave, and the slave responds after receiving the instruction from the host, and transmits the electric field strength detection result to the data processing unit through the modbusRTU transmission mode.

[0025] Furthermore, a preferred embodiment is proposed, wherein the data visualization unit comprises: a touch screen, a keyboard, and a mouse, wherein the touch screen is arranged on the rotating shaft of the handheld bracket.

[0026] Furthermore, a preferred method is proposed, in which the touch display screen is embedded with an rviz window, an OpenCV window and a PCL visualization window; the rviz window displays three-dimensional point cloud data and adjusts the display time of the point cloud; the OpenCV window displays processed image data; and the PCL visualization window displays the point cloud data of the laser radar.

[0027] Furthermore, a preferred method is proposed, in which the field of view of the camera is 69°*42°; the frame rate is 1920*1080; the range of the laser radar is 0.1~70m, and the ranging accuracy is 2cm.

[0028] Based on the same inventive concept, the present invention also proposes a visual detection device for the live state of a substation, the device comprising: a visible light camera, a laser radar, an electric field sensor, and an embedded development board;

[0029] The visible light camera, laser radar, electric field sensor, and embedded development board are installed on a handheld device and connected to the embedded development board via a network cable. The visible light camera is connected to the embedded development board via a USB, and the electric field sensor is connected to the embedded development board via a serial port.

[0030] The present invention is beneficial in that:

[0031] Traditional substation equipment operation and maintenance usually requires personnel to perform high-altitude operations, especially at high places on power towers and substation equipment, which not only increases the risk of operations, but may also cause fatigue and increased stress on personnel. In particular, manual contact detection requires personnel to climb to the vicinity of the equipment for detection, which poses a major safety hazard. The device proposed in the present invention realizes real-time and comprehensive monitoring of the status of substation equipment by integrating information collection units such as visible light cameras, lidars, and electric field sensors, eliminating the risks of manual high-altitude operations. Workers can complete inspections without having to climb up power towers or high-altitude equipment in person, greatly reducing safety hazards.

[0032] Traditional manual inspection methods have certain limitations, especially in some difficult-to-reach locations or when the equipment status is complex, the accuracy and efficiency of manual inspection are difficult to guarantee. In addition, manual inspection relies on the professional experience of on-site staff, which may result in some minor faults not being discovered in time. In the present invention, through the electric field sensor, the device can accurately detect the energized state of the equipment. Compared with traditional manual inspection methods, electric field sensors can provide higher detection accuracy, especially in high-voltage environments, and can ensure real-time monitoring of the energized state of electrical equipment, which helps to detect potential faults or anomalies in a timely manner and avoid more serious problems caused by equipment failures.

[0033] In traditional operation and maintenance, manual inspection is often relied on to obtain the operating status of the equipment. However, this method is difficult to achieve real-time and comprehensive monitoring of the equipment status, especially in high-voltage and live environments, where manual inspection is not only dangerous but also inefficient. The device designed in the present invention can collect equipment operation data in a multi-dimensional and all-round manner by combining multiple sensing technologies such as visible light cameras, laser radars and electric field sensors. Laser radars can provide accurate spatial data, help form a three-dimensional model of the equipment, perform morphological monitoring and spatial change analysis; visible light cameras can perform visual inspections and supplement detailed information that cannot be perceived by humans. The data processing unit can receive data from each sensor through the USB interface, Ethernet and Modbus RTU protocols, respectively, and perform efficient data fusion and processing. Through the data visualization unit, these data can be presented to the operation and maintenance personnel in an intuitive and easy-to-understand form, helping them to make quick judgments. Whether it is equipment status, fault warning or historical data analysis, it can be presented more clearly through visualization, thereby improving operation and maintenance efficiency and decision-making accuracy.

[0034] The invention is applied to the field of maintenance of substation equipment. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 This is a schematic diagram of the structure of the visual detection device for the live state of substation equipment according to the first embodiment;

[0036] Figure 2 This is a physical picture of the device for visually detecting the live state of a substation according to the eleventh embodiment;

[0037] Figure 3 This is a detection flow chart of implementation mode eleven. DETAILED DESCRIPTION

[0038] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments.

[0039] Implementation method 1, see Figure 1 The present embodiment is described below. The device for visually detecting the live state of a substation equipment in the present embodiment comprises:

[0040] Power module, information acquisition unit, data processing unit and data visualization unit;

[0041] The information collection unit includes: a visible light camera, a laser radar, and an electric field sensor;

[0042] The data processing unit receives data from the visible light camera through a USB interface, receives data from the laser radar through Ethernet, and receives data from the electric field sensor through a Modbus RTU communication protocol;

[0043] The data visualization unit is connected to the data processing unit;

[0044] The power supply module is used to provide electric energy to the information acquisition unit, the data processing unit and the data visualization unit.

[0045] Traditional substation equipment operation and maintenance usually requires personnel to perform high-altitude operations, especially at high altitudes on power towers and substation equipment, which not only increases operational risks, but may also cause fatigue and increased stress on personnel. In particular, manual contact detection requires personnel to climb to the vicinity of the equipment for detection, which poses a major safety hazard. The device proposed in this embodiment integrates information acquisition units such as visible light cameras, lidars, and electric field sensors to achieve real-time and comprehensive monitoring of the status of substation equipment, eliminating the risks of manual high-altitude operations. Staff can complete inspections without having to climb up power towers or high-altitude equipment in person, greatly reducing safety hazards.

[0046] Traditional manual inspection methods have certain limitations, especially in some difficult-to-reach locations or when the equipment status is complex, the accuracy and efficiency of manual inspection are difficult to guarantee. In addition, manual inspection relies on the professional experience of on-site staff, which may result in some minor faults not being discovered in time. In this embodiment, through the electric field sensor, the device can accurately detect the energized state of the equipment. Compared with traditional manual inspection methods, electric field sensors can provide higher detection accuracy, especially in high-voltage environments, and can ensure real-time monitoring of the energized state of electrical equipment, which helps to detect potential faults or anomalies in a timely manner and avoid more serious problems caused by equipment failures.

[0047] In traditional operation and maintenance, manual inspection is often relied on to obtain the operating status of the equipment. However, this method is difficult to achieve real-time and comprehensive monitoring of the equipment status, especially in high-voltage and live environments. Manual inspection is not only dangerous but also inefficient. The device designed in this embodiment can collect equipment operation data in a multi-dimensional and all-round manner by combining multiple sensing technologies such as visible light cameras, laser radars and electric field sensors. Laser radars can provide accurate spatial data to help form a three-dimensional model of the equipment, perform morphological monitoring and spatial change analysis; visible light cameras can perform visual inspections and supplement detailed information that cannot be perceived by humans. The data processing unit can receive data from each sensor through the USB interface, Ethernet and Modbus RTU protocols, respectively, and perform efficient data fusion and processing. Through the data visualization unit, these data can be presented to the operation and maintenance personnel in an intuitive and easy-to-understand form, helping them to make quick judgments. Whether it is equipment status, fault warning or historical data analysis, it can be presented more clearly through visualization, thereby improving operation and maintenance efficiency and decision-making accuracy.

[0048] Implementation method 2. This implementation method is a further limitation of the visual detection device for the energized state of the substation equipment described in implementation method 1. The visible light camera is used to collect image data of the substation equipment and transmit the information to the data processing unit; the laser radar is used to collect three-dimensional point cloud data of the substation equipment and transmit the point cloud data to the data processing unit.

[0049] In this embodiment, the visible light camera can obtain image data of the substation equipment, provide information such as the surface features and appearance status of the equipment, and facilitate the detection of the visual features of the equipment. By collecting three-dimensional point cloud data, the laser radar can accurately capture the spatial geometry of the equipment and its position and changes in three-dimensional space. It is not affected by lighting conditions and is suitable for accurate measurement in different environments. The combination of the two can achieve multi-level and multi-angle information fusion, improve the comprehensiveness and accuracy of detection, and make up for the shortcomings of a single sensor.

[0050] Implementation method three: This implementation method further limits the visual detection device for the live state of substation equipment described in implementation method one. The data processing unit is an embedded development board, and the YOLO target detection algorithm is embedded in the embedded development board.

[0051] As a data processing unit, embedded development boards usually have the characteristics of low power consumption and miniaturization, and can be embedded in the monitoring environment of substation equipment for real-time data processing. Compared with traditional PCs or servers, embedded development boards save more space and energy, and are suitable for deployment at equipment sites for long-term stable operation. Embedded systems usually have the characteristics of strong real-time performance and good anti-interference performance, which are very suitable for live status monitoring tasks of substation equipment that require efficient and stable operation.

[0052] The advantage of YOLO lies in its high speed and accuracy, especially when processing large-scale data streams, it can ensure a faster detection speed without sacrificing detection accuracy. This is crucial for real-time monitoring of substation equipment, and can obtain the operating status of the equipment in real time without interruption.

[0053] The YOLO target detection algorithm can be used to accurately analyze and process images of substation equipment, thereby identifying whether the equipment is energized, the damage of the equipment, or the fault point. By visually displaying the status of substation equipment, the working status of the equipment can be more intuitively understood, making it easier for operation and maintenance personnel to take timely measures.

[0054] Embodiment 4: This embodiment further limits the visual detection device for the live state of substation equipment described in Embodiment 3. The processing flow of the data processing unit is as follows:

[0055] The embedded development board receives the images collected by the visible light camera, the point cloud data acquired by the lidar, and the field strength data returned by the electric field sensor;

[0056] Perform target detection on the collected images, identify the substation equipment in the images, obtain the category information of the substation equipment, mark the substation equipment with a detection box, and calculate the pixel values ​​of the four vertices of the detection box;

[0057] Filter the collected point cloud data;

[0058] The point cloud data inside the detection frame is classified using Euclidean clustering, and the center point is fitted as the center point of the substation through the external bounding box method, and the distance of the substation is obtained through the center point;

[0059] The distance between a person and the substation is calculated using the electric field strength measured by the electric field sensor and the voltage of the substation, and compared with the distance to the substation obtained through the central point to determine the specific location and energized state of the energized equipment.

[0060] Implementation method 5. This implementation method further limits the visual detection device for the energized state of the substation equipment described in implementation method 1. The electric field sensor is used to continuously measure the AC power frequency electric field near the 110kV AC voltage line to obtain the electric field strength information radiated outward by the substation equipment.

[0061] Implementation method six. This implementation method is a further limitation of the visual detection device for the energized state of the substation equipment described in implementation method three. The electric field detection range of the electric field sensor is -50kV / m to 50kV / m, which is used to obtain the electric field strength of the substation equipment that radiates electric field outward from the surrounding area. The embedded development board is equivalent to the host in the modbus communication mechanism, and the electric field sensor is equivalent to the slave. The host sends a read instruction to the slave, and the slave responds after receiving the instruction from the host, and transmits the electric field strength detection result to the data processing unit through the modbusRTU transmission mode.

[0062] In this implementation, the communication between the embedded development board as the host and the electric field sensor as the slave can ensure the stability and efficiency of data transmission. The protocol supports the expansion of multiple devices and is suitable for application in complex substation environments, reducing the difficulty of system integration. The division of labor between the host and the slave is clear, and the host can centrally control and manage each electric field sensor, avoiding direct communication conflicts between multiple devices and further improving the stability of the system.

[0063] Furthermore, the electric field sensor has an electric field detection range of -50kV / m to 50kV / m, which can accurately capture the electric field strength around the substation equipment. This range covers the electric field strength of common power equipment, ensuring comprehensive monitoring of the equipment's energized state. The electric field sensor transmits the electric field strength detection results to the data processing unit in real time via the Modbus RTU protocol, allowing monitoring personnel to understand the energized state of the substation equipment in a timely manner, avoiding the delay and error of traditional manual inspections.

[0064] Implementation method 7: This implementation method further limits the device for visually detecting the live state of substation equipment described in implementation method 1. The data visualization unit includes: a touch screen, a keyboard, and a mouse, wherein the touch screen is arranged on the rotating shaft of the handheld bracket.

[0065] Implementation method eight. This implementation method is a further limitation of the visual detection device for the energized state of the substation equipment described in implementation method one. The touch-screen display is embedded with an rviz window, an OpenCV window and a PCL visualization window; three-dimensional point cloud data is displayed in the rviz window, and the display time of the point cloud is adjusted; processed image data is displayed in the OpenCV window; and point cloud data of the laser radar is displayed in the PCL visualization window.

[0066] In this embodiment, RViz is used to display three-dimensional point cloud data, which can help operators observe the spatial structure and status of substation equipment more intuitively. Three-dimensional point cloud data can usually provide complete spatial information of equipment, especially for places with complex structures or difficult equipment maintenance. Three-dimensional visualization helps to better judge the health of equipment, especially when detecting live state, it can avoid direct contact with live parts. By displaying image data through OpenCV, it is possible to display processed real-time images or video streams, which is convenient for detailed analysis of the surface of the equipment and detection of any abnormalities (such as equipment surface contamination, corrosion, cracks, etc.), so as to timely discover potential faults or safety hazards. Image data usually contains more texture, color and other information, which helps to assist in judging problems. LiDAR provides high-precision depth information, combined with PCL (Point Cloud Library) for point cloud processing, it can achieve accurate measurement and modeling of the surrounding environment. LiDAR point cloud data displayed through the PCL window can help monitor environmental changes around the equipment in real time, prevent the operating status of substation equipment from being affected by changes in the external environment, and is particularly suitable for detecting physical obstacles or environmental changes.

[0067] Implementation method nine. This implementation method is a further limitation of the visual detection device for the live state of substation equipment described in implementation method one. The field of view of the camera is 69°*42°; the frame rate is 1920*1080; the range of the laser radar is 0.1~70m, and the ranging accuracy is 2cm.

[0068] Embodiment 10: The visual detection device for the live state of the substation equipment described in this embodiment comprises: a visible light camera, a laser radar, an electric field sensor, and an embedded development board;

[0069] The visible light camera, laser radar, electric field sensor, and embedded development board are installed on a handheld device and connected to the embedded development board via a network cable. The visible light camera is connected to the embedded development board via a USB, and the electric field sensor is connected to the embedded development board via a serial port.

[0070] Implementation method 11: See Figure 1 and Figure 2 This embodiment provides a specific example of the visual detection device for the live state of the substation equipment described in the first embodiment, and is also used to explain the second to eighth embodiments. Specifically:

[0071] The entire device can be divided into a power supply, an information acquisition unit, a data processing unit and a data visualization unit.

[0072] The power supply uses a 12V DC power supply to power the embedded development board, lidar and electric field sensor, and the visible light camera is indirectly powered by connecting to the embedded development board.

[0073] The information collection unit includes a visible light camera module, a laser radar module, and an electric field sensor module. The visible light camera uses Intel's RealSense D435i camera, which has a field of view of 69° (H) * 42° (V) and a frame rate of up to 30FPS@1920*1080, and is used to capture image information of substation equipment in the surrounding environment. The visible light camera runs on the ROS robot operating system, publishes the collected image data as the topic / camera / color / image_raw, and transmits it to the data processing unit via USB3.0. The laser radar uses Livox's Mid-360 laser radar, which has a range of 0.1 to 70m and a ranging accuracy of 2cm. It can effectively ensure the accuracy and efficiency of the spatial posture data measurement of substation equipment. The laser radar is used to collect three-dimensional spatial point cloud data of substation equipment within the range. Similarly, the laser radar also runs on the ROS robot operating system, publishes the collected point cloud data as the topic / livox / lidar, and transmits it to the data processing unit via Ethernet. The electric field sensor has an electric field detection range of -50kV / m to 50kV / m, and is used to obtain the electric field strength of the surrounding substation equipment that radiates electric fields outward. The embedded development board is equivalent to the host in the modbus communication mechanism, and the electric field sensor is equivalent to the slave. The host sends a read instruction to the slave, and the slave responds after receiving the instruction from the host, and transmits the electric field strength detection result to the data processing unit through the modbusRTU transmission mode.

[0074] The hardware part of the data processing unit uses the NVIDIA Jetson Xavier NX embedded development board, which provides 21TOPS computing power and supports USB, Ethernet and modbus communications. The software part includes the Ubuntu18.04 operating system and the ROS robot operating system as well as the C++ program written by the user running on ROS. The user program receives the image and point cloud data of the visible light camera and the lidar by subscribing to the ROS topic published by the information acquisition unit, and then processes the image, point cloud and electric field strength data transmitted by the information acquisition unit, and sends the processed data to the data visualization unit for intuitive display of the detection results. The data processing program includes the YOLO target detection algorithm for processing image data, the point cloud processing program for realizing the three-dimensional spatial recognition of the substation equipment, and the electric field strength processing program for judging whether the substation equipment is energized. The YOLO target detection algorithm has been rewritten in C++ and accelerated by TensorRT, making the detection speed of the substation equipment detection model faster than the original PyTorch model and consuming less computing resources.

[0075] The hardware part of the data visualization unit includes a touch screen, keyboard and mouse. The display screen can be rotated up and down through the shaft of the handheld bracket to obtain the best observation angle. The software part includes rviz, OpenCV and PCL visualization windows, which are used to intuitively display the data processed by the data processing unit. The decay time of the point cloud data can be adjusted in the rviz window to make the point cloud effect clearer, which is convenient for operators to observe the position, spatial posture and energized state of the substation equipment.

[0076] The following is the detailed operation process of this device. Figure 3 As shown:

[0077] (1) According to the attached Figure 1 Build the hardware platform and install the visible light camera, lidar, electric field sensor, and embedded development board on the handheld device. Connect the lidar to the embedded development board via a network cable, the visible light camera to the embedded development board via USB, and the electric field sensor to the embedded development board via a serial port.

[0078] (2) Turn on the embedded development board, the lidar, and the visible light camera so that the lidar and the visible light camera can collect data, the embedded development board can process the data, the electric field sensor can send data, and the display can display the results.

[0079] (3) The embedded development board receives the image captured by the visible light camera, a frame of point cloud data obtained by the lidar, and the field strength data returned by the electric field sensor.

[0080] (4) Performing target detection on the captured image, identifying the substation equipment in the image, obtaining the category information of the substation equipment, marking the substation equipment with a detection box, and calculating the pixel values ​​of the four vertices of the detection box.

[0081] (5) Filter the collected point cloud data. First, perform voxel filtering and downsampling to remove a certain degree of noise points without losing the shape characteristics of the detected object. Use a straight-through filter to extract the region of interest and segment the ground point cloud data. Finally, eliminate outliers to prepare for subsequent clustering.

[0082] (6) Use Euclidean clustering to classify the point cloud data inside the detection box, and use the external bounding box method to fit the center point as the center point of the substation equipment. The distance and other spatial posture data of the substation equipment are obtained through the center point.

[0083] The distance between the person and the substation is calculated using the electric field strength measured by the electric field sensor and the voltage of the substation, and compared with the distance between the laser radar and the substation measured in step (6) to determine the specific location and energized state of the energized equipment.

[0084] The final test results are visualized on the display screen.

[0085] The present invention designs a handheld device for visually detecting the live state of a transformer, as shown in the figure. Figure 2 The invention realizes the long-distance and non-contact detection of the live state of the substation equipment, reduces the safety risk brought by the close-distance contact detection to the operation and maintenance personnel, and has the advantages of higher safety and better usability compared with the existing manual contact detection method.

[0086] Although the preferred embodiments of the present disclosure have been described, those skilled in the art may make additional changes and modifications to these embodiments once they have learned the basic creative concept. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments and all changes and modifications falling within the scope of the present disclosure.

[0087] Obviously, those skilled in the art can make various changes and modifications to the present disclosure without departing from the spirit and scope of the present disclosure. Thus, if these modifications and variations of the present disclosure fall within the scope of the claims of the present disclosure and their equivalents, the present disclosure is also intended to include these modifications and variations.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present disclosure rather than to limit its protection scope. Although the present disclosure has been described in detail with reference to the above embodiments, ordinary technicians in the relevant field should understand that after reading the present disclosure, those skilled in the art can still make various changes, modifications or equivalent substitutions to the specific implementation methods of the invention, but these changes, modifications or equivalent substitutions are all within the protection scope of the disclosed claims to be approved.

Claims

1. A visual detection device for the live state of substation equipment, characterized in that: The device comprises: Power module, information acquisition unit, data processing unit and data visualization unit; The information collection unit includes: a visible light camera, a laser radar, and an electric field sensor; The data processing unit receives data from the visible light camera via a USB interface, receives data from the laser radar via Ethernet, and receives data from the electric field sensor via a Modbus RTU communication protocol; The data visualization unit is connected to the data processing unit; The power supply module is used to provide electric energy to the information acquisition unit, the data processing unit and the data visualization unit.

2. The device for visually detecting the live state of a power transformation equipment according to claim 1, characterized in that: The visible light camera is used to collect image data of the substation equipment and transmit the information to the data processing unit; the laser radar is used to collect three-dimensional space point cloud data of the substation equipment and transmit the point cloud data to the data processing unit.

3. The device for visually detecting the live state of a power transformation equipment according to claim 1, characterized in that: The data processing unit is an embedded development board, and the embedded development board has a YOLO target detection algorithm embedded therein.

4. The device for visually detecting the live state of a power transformation equipment according to claim 3, characterized in that: The processing flow of the data processing unit is as follows: The embedded development board receives images captured by the visible light camera, point cloud data acquired by the lidar, and field intensity data returned by the electric field sensor; Perform target detection on the captured image, identify the substation equipment in the image, obtain the category information of the substation equipment, mark the substation equipment with a detection box, and calculate the pixel values ​​of the four vertices of the detection box; Filter the collected point cloud data; Use Euclidean clustering to classify the point cloud data inside the detection box, and use the external bounding box method to fit the center point as the center point of the substation equipment, and obtain the distance of the substation equipment through the center point; The distance between the person and the substation is calculated using the electric field strength measured by the electric field sensor and the voltage of the substation. This is then compared with the distance to the substation obtained through the center point to determine the specific location and energized status of the energized equipment.

5. The device for visually detecting the live state of a power transformation equipment according to claim 1, characterized in that: The electric field sensor is used to continuously measure the AC power frequency electric field near the 110kV AC voltage line and obtain the electric field strength information radiated outward by the substation equipment.

6. The device for visually detecting the live state of a power transformation equipment according to claim 3, characterized in that: The electric field sensor has an electric field detection range of -50kV / m to 50kV / m and is used to obtain the electric field strength of the surrounding substation equipment that radiates electric fields outward. The embedded development board is equivalent to the host in the Modbus communication mechanism, and the electric field sensor is equivalent to the slave. The host sends a read instruction to the slave, and the slave responds after receiving the instruction from the host and transmits the electric field strength detection result to the data processing unit through the Modbus RTU transmission mode.

7. The device for visually detecting the live state of a power transformation equipment according to claim 1, characterized in that: The data visualization unit includes: a touch screen, a keyboard, and a mouse, wherein the touch screen is arranged on the rotating shaft of the handheld bracket.

8. The device for visually detecting the live state of a power transformation equipment according to claim 1, characterized in that: The touch screen display is embedded with an rviz window, an OpenCV window, and a PCL visualization window; the rviz window displays three-dimensional point cloud data and adjusts the display time of the point cloud; the OpenCV window displays processed image data; The PCL visualization window displays the point cloud data of the laser radar.

9. The device for visually detecting the live state of a power transformation equipment according to claim 1, characterized in that: The camera's field of view is 69°*42°; the frame rate is 1920*1080; the laser radar has a range of 0.1 to 70 meters and a ranging accuracy of 2 cm.

10. A visual detection device for the live state of a power substation, characterized in that: The device includes: a visible light camera, a laser radar, an electric field sensor, and an embedded development board; The visible light camera, laser radar, electric field sensor, and embedded development board are installed on a handheld device and connected to the embedded development board via a network cable. The visible light camera is connected to the embedded development board via a USB, and the electric field sensor is connected to the embedded development board via a serial port.

Images