Method and device for detecting high-voltage GIS (Gas Insulated Switchgear) equipment

Abstract

The invention provides a detection method and device for high-voltage GIS equipment, and relates to the technical field of detection. The method comprises the following steps: acquiring a transformer substation environment image containing high-voltage GIS equipment and registered three-dimensional point cloud data corresponding to the transformer substation environment; determining a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area comprises a key part of the high-voltage GIS equipment; and verifying the candidate target area according to a pre-established GIS equipment three-dimensional model to obtain a target area position conforming to equipment characteristics, and controlling an execution mechanism to detect the target area position. The device executes the method. According to the detection method and device of the high-voltage GIS equipment provided by the embodiment of the invention, the safety and efficiency of the detection of the high-voltage GIS equipment can be improved.

Description

Technical Field

[0001] The present invention relates to the field of detection technology, and in particular to a detection method and device for high-voltage GIS equipment. Background Art

[0002] In the complex and hazardous operation scenarios of the power industry, new robot technology has huge application potential and demand. The maintenance technology, safety assurance and intelligent requirements of power equipment are gradually improving, and the application of new robot technology in this field will be more extensive and in-depth. However, this field often faces harsh operating environments and multiple potential dangerous factors. Equipment maintenance and overhaul work requires highly specialized personnel to carry out in a high-risk environment, which poses a series of technical and operational challenges such as personal safety hazards and low efficiency. Summary of the invention

[0003] In view of the problems in the prior art, an embodiment of the present invention provides a detection method and device for high-voltage GIS equipment, which can at least partially solve the problems in the prior art.

[0004] On the one hand, the present invention provides a detection method for high-voltage GIS equipment, comprising:

[0005] Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment;

[0006] Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment;

[0007] The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

[0008] Wherein, obtaining the three-dimensional point cloud data includes:

[0009] Preprocess the initial three-dimensional point cloud data of the power station environment obtained by LiDAR;

[0010] Insert the preprocessed initial 3D point cloud data into the local map, and mark the 3D point cloud data inserted into the local map as active;

[0011] Perform scene registration on the three-dimensional point cloud data in an active state to obtain the three-dimensional point cloud data.

[0012] The step of performing scene registration on the active three-dimensional point cloud data to obtain the three-dimensional point cloud data includes:

[0013] Based on the SC2-PCR algorithm, the directional similarity measurement is calculated for the active three-dimensional point cloud data with preliminary correspondence;

[0014] The three-dimensional point cloud data whose direction similarity measurement calculation results are higher than a threshold value are determined as a first seed point set with high reliability, and the remaining three-dimensional point cloud data are determined as a second seed point set with low reliability;

[0015] Collectively sampling the first seed point set and the second seed point set according to a preset ratio, and performing an overall consistency evaluation to obtain an optimal transformation matrix;

[0016] The three-dimensional point cloud data in the active state is registered and calculated according to the optimal transformation matrix to obtain the three-dimensional point cloud data.

[0017] Wherein, obtaining the substation environment image includes:

[0018] Performing initial preprocessing on the initial substation environment image including high-voltage GIS equipment obtained by the depth camera;

[0019] The initial substation environment image after the primary preprocessing is used as the substation environment image.

[0020] Wherein, determining the candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data includes:

[0021] Performing secondary preprocessing on the substation environment image to obtain an optimized substation environment image;

[0022] Acquire edge features of the high-voltage GIS equipment according to the optimized substation environment image;

[0023] Performing secondary preprocessing on the three-dimensional point cloud data to obtain optimized three-dimensional point cloud data;

[0024] Acquire key geometric features of the high-voltage GIS equipment according to the optimized three-dimensional point cloud data;

[0025] The edge features and the key geometric features are fused, and the candidate target area is determined according to the feature fusion result.

[0026] The control execution mechanism detects the position of the target area, including:

[0027] Sending a control instruction to a mechanical arm mounted on the UAV so that the mechanical arm detects the position of the target area according to the control instruction;

[0028] Wherein, the control instruction carries the target area position information corresponding to the target area position; the drone is also equipped with a laser radar and a depth camera.

[0029] Wherein, the detection method of the high-voltage GIS equipment further includes:

[0030] If it is determined that the positioning of the UAV fails during the process of inserting the pre-processed initial three-dimensional point cloud data into the local map, the UWB device is used to reposition the UAV.

[0031] On the one hand, the present invention provides a detection device for high-voltage GIS equipment, comprising:

[0032] An acquisition unit, used to acquire an environmental image of a substation including high-voltage GIS equipment, and registered three-dimensional point cloud data of the substation environment;

[0033] A determination unit, configured to determine a candidate target area for detecting the high-voltage GIS device according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS device;

[0034] The detection unit is used to verify the candidate target area according to the pre-established three-dimensional model of the GIS equipment, obtain the target area position that meets the equipment characteristics, and control the actuator to detect the target area position.

[0035] In another aspect, an embodiment of the present invention provides a computer device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the following method when executing the computer program:

[0036] Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment;

[0037] Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment;

[0038] The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

[0039] An embodiment of the present invention provides a computer-readable storage medium, including:

[0040] The computer readable storage medium stores a computer program, and when the computer program is executed by a processor, the following method is implemented:

[0041] Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment;

[0042] Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment;

[0043] The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

[0044] The detection method and device of high-voltage GIS equipment provided in the embodiment of the present invention obtain a substation environment image including the high-voltage GIS equipment, and three-dimensional point cloud data after registration of the substation environment; determine a candidate target area for detecting the high-voltage GIS equipment based on the substation environment image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS equipment; verify the candidate target area based on a pre-established three-dimensional model of the GIS equipment, obtain the target area position that meets the equipment characteristics, and control the actuator to detect the target area position, which can improve the safety and efficiency of high-voltage GIS equipment detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings:

[0046] Figure 1 It is a flow chart of a detection method for high-voltage GIS equipment provided in one embodiment of the present invention.

[0047] Figure 2 It is a structural schematic diagram of a detection device for high-voltage GIS equipment provided by an embodiment of the present invention.

[0048] Figure 3 A schematic diagram of the physical structure of a computer device provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] In order to make the purpose, technical scheme and advantages of the embodiments of the present invention more clear, the embodiments of the present invention are further described in detail below in conjunction with the accompanying drawings. Here, the illustrative embodiments of the present invention and their descriptions are used to explain the present invention, but are not intended to limit the present invention. It should be noted that, in the absence of conflict, the embodiments in this application and the features in the embodiments can be combined with each other arbitrarily.

[0050] The professional terms of the embodiments of the present invention are described as follows:

[0051] GIS: stands for "Gas Insulated Switchgear", which is a key equipment used in high-voltage power systems. GIS usually consists of circuit breakers, disconnectors, earthing switches, current transformers and voltage transformers, etc. It uses insulating gas (usually sulfide gas) to isolate and insulate high-voltage components to ensure safe and reliable operation of the equipment.

[0052] R3LIVE algorithm: By integrating lidar and visual data, it achieves robust and accurate positioning. Its key innovation is to take advantage of multi-sensor data and combine it with advanced feature extraction and optimization techniques to ensure high-precision real-time positioning and mapping in complex environments. This makes R3LIVE particularly suitable for drone live detection tasks in high-voltage GIS equipment substations, significantly improving the safety, efficiency and accuracy of detection.

[0053] The SC2-PCR (Scene Consistency-based Point Cloud Registration) algorithm achieves high-precision and robust point cloud registration by evaluating the directional similarity and consistency in point cloud data. Its key terms include: point cloud (data consisting of 3D coordinate points), directional similarity measure (evaluating the consistency of point clouds in different directions), seed points (high-confidence point pairs for preliminary registration), co-collective sampling (assessing overall consistency by randomly sampling point pairs) and registration (aligning different point cloud data to the same coordinate system). These terms together describe how the algorithm achieves accurate point cloud alignment in complex scenes.

[0054] Figure 1 FIG. 1 is a flow chart of a detection method for high-voltage GIS equipment provided by an embodiment of the present invention. Figure 1 As shown, the detection method of high-voltage GIS equipment provided by the embodiment of the present invention includes:

[0055] Step S1: Acquire an environmental image of a substation including high-voltage GIS equipment, and registered three-dimensional point cloud data of the substation environment.

[0056] Step S2: determining a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS equipment.

[0057] Step S3: verify the candidate target area according to the pre-established three-dimensional model of the GIS equipment, obtain the target area position that meets the equipment characteristics, and control the actuator to detect the target area position.

[0058] In the above step S1, the device obtains an image of a substation environment including a high-voltage GIS device, and three-dimensional point cloud data of the substation environment after registration. The device can be a computer device that executes the method, such as a server. The acquisition, storage, use, and processing of data in the technical solution of this application are in compliance with relevant regulations.

[0059] First, the data collection is described as follows:

[0060] LiDAR data collection: Scan the substation environment with LiDAR to obtain initial 3D point cloud data.

[0061] Depth camera data acquisition: Use a depth camera to photograph substation equipment and obtain color images and their corresponding depth information, that is, the initial substation environment image containing high-voltage GIS equipment.

[0062] IMU inertial data acquisition: Use the inertial measurement unit (IMU) to record the motion data of the drone, including acceleration and angular velocity.

[0063] UWB positioning data collection: UWB (ultra-wideband) equipment obtains the initial position and attitude information of the drone, uses ultra-wideband signals for high-precision ranging and positioning, and calculates the position of the drone by measuring the distance between the tag and multiple anchor points.

[0064] Acquiring the three-dimensional point cloud data includes:

[0065] Preprocess the initial three-dimensional point cloud data of the power station environment obtained by LiDAR;

[0066] Insert the preprocessed initial 3D point cloud data into the local map, and mark the 3D point cloud data inserted into the local map as active;

[0067] Perform scene registration on the three-dimensional point cloud data in an active state to obtain the three-dimensional point cloud data.

[0068] The R3LIVE (Real-time Robust and Resilient LiDAR-Inertial-Visual Odometry) algorithm achieves real-time, high-precision positioning and map construction by fusing data from LiDAR, inertial measurement unit (IMU) and depth camera.

[0069] The execution process is as follows: The system first waits for the initial 3D point cloud data with the latest timestamp in the LIO (LiDAR-Inertial Odometry) module and preprocesses it to ensure the validity and reliability of the data. After obtaining the initial 3D point cloud data, the motion information provided by the IMU is used to dedistort the point cloud data to correct the point cloud distortion caused by sensor motion. Then, the extended state Kalman filter (ESIKF) prediction is performed based on the IMU data:

[0070] The state prediction formula is as follows:

[0071]

[0072] is the predicted value of the state estimate at time k-1.

[0073] f is the system state transfer function, which describes the state change of the system between time k-1 and time k.

[0074] is the state estimate at time k-1.

[0075] μ k is the control input at time k.

[0076] Covariance prediction:

[0077]

[0078] P k|k-1 is the predicted covariance matrix of the state estimation error.

[0079] F k-1 is the Jacobian matrix of the state transfer matrix, which describes the linearization of the state estimate by the state transfer function.

[0080] P k-1|k-1 is the state estimation error covariance matrix at time k-1.

[0081] Q K-1 is the covariance matrix of the process noise, describing the uncertainty or modeling error of the system.

[0082] The high-frequency data provided by the IMU is used to update the system's predictions, including position, velocity, and attitude. Then, combined with the processed laser point cloud data, the ESIKF state is updated to further correct and optimize the system state and improve positioning accuracy. Subsequently, the processed laser point cloud data is inserted into the local map, and the point clouds are marked as active for subsequent use and updates. Throughout the process, the system's state quantities (including position, attitude, etc.) are continuously updated and maintained to ensure accurate positioning and stable operation of the system. The system maintains a point cloud map that records all laser point cloud data and their location information to achieve environmental reconstruction and continuous optimization of positioning.

[0083] The performing scene registration on the active three-dimensional point cloud data to obtain the three-dimensional point cloud data includes:

[0084] Based on the SC2-PCR algorithm, the directional similarity measurement is calculated for the active three-dimensional point cloud data with preliminary correspondence;

[0085] The three-dimensional point cloud data whose direction similarity measurement calculation results are higher than a threshold value are determined as a first seed point set with high reliability, and the remaining three-dimensional point cloud data are determined as a second seed point set with low reliability;

[0086] Collectively sampling the first seed point set and the second seed point set according to a preset ratio, and performing an overall consistency evaluation to obtain an optimal transformation matrix;

[0087] The three-dimensional point cloud data in the active state is registered and calculated according to the optimal transformation matrix to obtain the three-dimensional point cloud data.

[0088] The SC2-PCR (Scene Consistency-based Point Cloud Registration) algorithm performs scene registration for the collected point cloud data (active 3D point cloud data). Its innovation lies in achieving high-precision and robust point cloud registration through a series of refined steps. The specific execution process is as follows:

[0089] 1. Input of initial correspondence:

[0090] First, the algorithm receives active 3D point cloud data and its preliminary correspondences, which can be obtained through feature matching or other preliminary alignment methods.

[0091] 2. Calculate the directional similarity measure (SC2):

[0092] The algorithm calculates the similarity measure (SC2) in each direction of the active 3D point cloud data with preliminary correspondence. This process evaluates the consistency and similarity of the point cloud in different directions, thus preliminarily screening out possible alignment directions.

[0093]

[0094] SC2 IJ Represents the directional similarity measure of the point pair (i, j).

[0095] n i and n j Are the normal vectors of point i and point j respectively.

[0096] · represents the vector dot product operation.

[0097] ||n i || and ||n j || are the magnitudes of the normal vectors of i and j respectively.

[0098] 3. Select the first set of seed points with high reliability and the second set of seed points with low reliability:

[0099] The algorithm selects a set of reliable seed points as the basis for the subsequent registration process. These seed points are high-confidence point pairs selected from the similarity measures calculated in the previous step, ensuring the accuracy and credibility of their correspondence between different point clouds. Seed selection conditions:

[0100] If SC2 IJ > threshold, then (i, j) ∈ seeds

[0101] where threshold corresponds to the threshold, and the set of seed points with SC2 IJ = <threshold is used as the second set of seed points.

[0102] 4. Consensus Sampling:

[0103] The algorithm performs Consensus Sampling in the two sets of seed points according to a preset ratio. This step further improves the robustness and accuracy of the registration by randomly selecting point pairs and evaluating their overall consistency. The preset ratio can be set independently according to the actual situation, and the optional ratio is the first set of seed points: the second set of seed points = 7:3.

[0104] Steps of Consensus Sampling:

[0105] Randomly select N pairs of point pairs (p k , q k);

[0106] Calculate the transformation matrix T for each pair of points K , which can be calculated using existing technology.

[0107] Evaluate all T K The overall consistency of the transformation matrix T is selected. * To achieve overall consistency evaluation, the expression is as follows:

[0108]

[0109] T K Represents the point pair (p k ,q k ) to calculate the transformation matrix.

[0110] T * represents the optimal transformation matrix.

[0111] ||·|| represents the Euclidean norm of a vector.

[0112] 5. Transformation matrix application and registration completed:

[0113] Select the optimal transformation matrix T * Finally, it is applied to the point cloud data to complete the registration.

[0114] Transformation application formula:

[0115] P′=T * ·P

[0116] P is the active 3D point cloud data.

[0117] P′ is the registered three-dimensional point cloud data of the strain power station environment.

[0118] Acquiring the substation environment image, including:

[0119] Performing initial preprocessing on the initial substation environment image including high-voltage GIS equipment obtained by the depth camera;

[0120] The initial substation environment image after the primary preprocessing is used as the substation environment image.

[0121] At the VIO (Visual-Inertial Odometry) terminal, the system waits for the initial substation environment image with the latest timestamp containing the high-voltage GIS equipment and pre-processes it. The IMU data is used to predict the ESIKF state and update the camera status information.

[0122] Next, the optical flow method is used to track the feature points in the previous frame to ensure that the position of the feature points in the current frame is accurately predicted. The initial pose of the matching feature points is calculated using the PNP algorithm (Perspective-n-Point) to provide initial values ​​for subsequent optimization.

[0123] Then, combined with the camera’s measurement data, the ESIKF state is updated to further correct the system state. The updated state is fed back to the IMU module to further correct the IMU state prediction. Color information is extracted from the camera image and the point cloud data color is updated through Bayesian.

[0124] In the above step S2, the device determines a candidate target area for detecting the high-voltage GIS equipment based on the substation environment image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS equipment. The determination of the candidate target area for detecting the high-voltage GIS equipment based on the substation environment image and the three-dimensional point cloud data includes:

[0125] Performing secondary preprocessing on the substation environment image to obtain an optimized substation environment image;

[0126] Acquire edge features of the high-voltage GIS equipment according to the optimized substation environment image;

[0127] Performing secondary preprocessing on the three-dimensional point cloud data to obtain optimized three-dimensional point cloud data;

[0128] Acquire key geometric features of the high-voltage GIS equipment according to the optimized three-dimensional point cloud data;

[0129] The edge features and the key geometric features are fused, and the candidate target area is determined according to the feature fusion result.

[0130] The substation environment image obtained by the depth camera and the three-dimensional point cloud data registered by the SC2-PCR algorithm are fused for data processing. The visual data provided by the depth camera has rich detail information, while the three-dimensional point cloud data provides accurate geometric shape information. In order to improve the quality of the image, image preprocessing (secondary image preprocessing) is performed, including denoising, contrast enhancement and brightness adjustment. Subsequently, the color image is converted into a grayscale image to simplify data processing and improve the accuracy of edge detection.

[0131] Next, the Canny edge detection algorithm is used to perform edge detection on the preprocessed grayscale image. The Canny edge detection algorithm calculates the image gradient and finds significant edge features in the image, thereby extracting the outline and edge of the GIS equipment. The specific steps include: first, Gaussian filtering the image to smooth the image and reduce the impact of noise on edge detection; then calculating the gradient amplitude and direction of the image to identify the edge pixels in the image; refining the gradient amplitude through non-maximum suppression to retain the most significant edge pixels; finally, setting high and low thresholds, determining strong and weak edges through dual threshold detection, and connecting the edges to form a complete edge map.

[0132] At the same time, the 3D point cloud data after SC2-PCR algorithm registration is processed (secondary preprocessing of point cloud data). First, the point cloud data is segmented to extract the point cloud area that may contain GIS equipment. Then, feature extraction is performed on the extracted point cloud area to identify points with significant geometric features (key geometric features), such as planes, corners, and edges. The edge features extracted from the image are matched and fused with the key geometric features, and the accuracy and robustness of target detection and positioning are improved by combining data from multiple sensors.

[0133] The candidate target area can be determined according to the feature fusion result, which can be achieved by using existing mature technologies and will not be described in detail.

[0134] In the above step S3, the device verifies the candidate target area according to the pre-established three-dimensional model of the GIS equipment, obtains the target area position that meets the equipment characteristics, and controls the actuator to detect the target area position. These candidate target areas may contain key parts of the GIS equipment. Using the pre-established three-dimensional model of the GIS equipment, the candidate target areas are verified and compared, and the target area positions that best meet the equipment characteristics are screened out. Then, combined with multi-frame images and point cloud data, the position is accurately processed to eliminate noise and errors, and finally the key parts of the GIS equipment are determined. The target area position information including the three-dimensional coordinates of the target area position (relative to the global coordinate system of the substation environment) is also determined at this stage.

[0135] The control execution mechanism detects the position of the target area, including:

[0136] Sending a control instruction to a mechanical arm mounted on the UAV so that the mechanical arm detects the position of the target area according to the control instruction;

[0137] Wherein, the control instruction carries the target area position information corresponding to the target area position; the drone is also equipped with a laser radar and a depth camera.

[0138] The detection method of the high-voltage GIS equipment also includes:

[0139] If the positioning of the drone is determined to be invalid during the process of inserting the pre-processed initial 3D point cloud data into the local map, the UWB device is used to reposition the drone. Under extreme conditions where the SLAM (Simultaneous Localization and Mapping) odometer fails, the UWB device is used for repositioning.

[0140] UWB devices use the time difference positioning technology of radio waves to provide additional location information, ensuring the continuous positioning capability of drones in complex environments.

[0141] The detection method of high-voltage GIS equipment provided by the embodiment of the present invention can specifically solve the following technical problems:

[0142] 1: Often faced with harsh working environments and multiple potential risk factors, equipment maintenance and overhaul work requires highly specialized personnel to carry out in a high-risk environment, which poses a risk of personal safety hazards.

[0143] 2: It relies on the experience, skills and physical strength of the inspectors. The inspection time is long, the types and quantity of equipment to be inspected are large, the inspection speed is slow, and due to the heavy workload and long time, the inspectors are prone to fatigue, resulting in reduced inspection efficiency.

[0144] 3: Due to the inherent limitations of human eyes, inspection with a high-power magnifying glass or microscope is also subject to subjective influence, making accuracy impossible to ensure. It is necessary to manually record and process a large amount of data, which is prone to errors and inefficiency. The standards of different inspectors may vary, which is easy to detect errors and miss inspections.

[0145] 4: Traditional GIS equipment detection methods require power outages, which not only affects power supply but also poses certain safety risks.

[0146] 5: In most solutions, drones are mainly used for patrol navigation in substations, and there are still deficiencies in equipment identification and live detection in substations.

[0147] Through UAV and multi-sensor fusion technology, an efficient, safe and accurate live detection method for high-voltage GIS equipment in substations is provided to solve the safety, efficiency and accuracy problems of traditional detection methods, realize intelligent and automated detection of substation equipment, and improve the operational reliability and maintenance level of the power system.

[0148] The detection method of high-voltage GIS equipment provided by the embodiment of the present invention has the following beneficial technical effects:

[0149] Application of R3LIVE algorithm: The integration of LiDAR and visual data achieves robust and accurate positioning, improving the navigation capability of drones in substation environments.

[0150] UWB device relocalization: Using UWB devices for relocalization under extreme conditions ensures that location information can still be obtained when the SLAM odometer fails, thus enhancing the robustness of the system.

[0151] C++ implementation of SC2-PCR algorithm: The SC2-PCR algorithm is used for scene registration, which has strong robustness and can achieve good registration effect even if the scene changes. Using C++ to implement the algorithm reduces dependence on the Python environment and is more convenient for engineering applications.

[0152] Target positioning combined with Canny edge detection: After scene registration, the target position of the GIS equipment is located through the Canny edge detection algorithm, combined with 3D point cloud data to achieve efficient and accurate live detection and status assessment.

[0153] Detection speed: Using intelligent inspection drones, multiple devices can be quickly scanned and inspected in a short period of time, greatly improving detection efficiency.

[0154] Detection accuracy: Through advanced sensors and image processing technology, the device status can be identified more accurately, reducing misjudgments and missed judgments. In terms of accuracy, machine detection has a general advantage in accuracy.

[0155] Adaptability to the working environment: Intelligent inspection drone equipment is not restricted by highly dangerous environmental conditions and can work stably in harsh environments to ensure detection efficiency and quality.

[0156] Continuous working ability: Intelligent inspection drones can work continuously without being affected by fatigue, ensuring detection efficiency. Data is processed quickly and accurately, reducing the time and workload of data processing and improving overall efficiency.

[0157] The detection method of high-voltage GIS equipment provided in the embodiment of the present invention obtains a substation environment image including the high-voltage GIS equipment and three-dimensional point cloud data after registration of the substation environment; determines a candidate target area for detecting the high-voltage GIS equipment based on the substation environment image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS equipment; verifies the candidate target area based on a pre-established three-dimensional model of the GIS equipment to obtain a target area position that meets the equipment characteristics, and controls an actuator to detect the target area position, thereby improving the safety and efficiency of high-voltage GIS equipment detection.

[0158] Further, obtaining the three-dimensional point cloud data includes:

[0159] The initial three-dimensional point cloud data of the strain power station environment obtained by the laser radar is preprocessed; the above-mentioned embodiment can be referred to for description and will not be repeated here.

[0160] The preprocessed initial three-dimensional point cloud data is inserted into the local map, and the three-dimensional point cloud data inserted into the local map is marked as active; the above-mentioned embodiment can be referred to for description and will not be repeated here.

[0161] The three-dimensional point cloud data in the active state is subjected to scene registration to obtain the three-dimensional point cloud data. The above-mentioned embodiments can be referred to for explanation and will not be described in detail.

[0162] Furthermore, performing scene registration on the active three-dimensional point cloud data to obtain the three-dimensional point cloud data includes:

[0163] Based on the SC2-PCR algorithm, the directional similarity measurement calculation is performed on the three-dimensional point cloud data in the active state with the preliminary corresponding relationship; the description can be made with reference to the above embodiment and will not be repeated here.

[0164] The three-dimensional point cloud data whose direction similarity measurement calculation results are higher than the threshold are determined as the first seed point set with high reliability, and the remaining three-dimensional point cloud data are determined as the second seed point set with low reliability; please refer to the above embodiment for description and will not repeat them here.

[0165] The first seed point set and the second seed point set are collectively sampled according to a preset ratio, and an overall consistency evaluation is performed to obtain an optimal transformation matrix; the above-mentioned embodiment can be referred to for description and will not be repeated here.

[0166] The three-dimensional point cloud data in the active state is registered and calculated according to the optimal transformation matrix to obtain the three-dimensional point cloud data. The above-mentioned embodiments can be referred to for explanation and will not be described in detail.

[0167] Further, obtaining the substation environment image includes:

[0168] The initial substation environment image including the high-voltage GIS equipment obtained by the depth camera is pre-processed for the first time; the above-mentioned embodiment can be referred to for description and will not be repeated here.

[0169] The initial substation environment image after the first preprocessing is used as the substation environment image. The above description can be referred to in the above embodiment and will not be repeated here.

[0170] Furthermore, the determining of a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data includes:

[0171] The substation environment image is subjected to secondary preprocessing to obtain an optimized substation environment image; the above-mentioned embodiment can be referred to for description and will not be described in detail.

[0172] The edge features of the high-voltage GIS equipment are obtained based on the optimized substation environment image; the above-mentioned embodiment can be referred to for description and will not be repeated here.

[0173] The three-dimensional point cloud data is subjected to secondary preprocessing to obtain optimized three-dimensional point cloud data; the above description may be referred to in the above embodiment and will not be repeated herein.

[0174] The key geometric features of the high-voltage GIS equipment are obtained based on the optimized three-dimensional point cloud data; please refer to the above-mentioned embodiment for description and no further details will be given.

[0175] The edge features and the key geometric features are fused, and the candidate target region is determined according to the feature fusion result. The above embodiments can be referred to for explanation, and no further description is given.

[0176] Furthermore, the control execution mechanism detects the position of the target area, including:

[0177] Send a control instruction to the mechanical arm mounted on the UAV so that the mechanical arm detects the position of the target area according to the control instruction; refer to the above embodiment for description and no further details will be given.

[0178] The control instruction carries the target area position information corresponding to the target area position; the drone is also equipped with a laser radar and a depth camera. The above-mentioned embodiments can be referred to for explanation and will not be described in detail.

[0179] Furthermore, the detection method of the high-voltage GIS equipment also includes:

[0180] If it is determined that the positioning of the UAV fails during the process of inserting the pre-processed initial three-dimensional point cloud data into the local map, the UWB device is used to reposition the UAV. The above-mentioned embodiments can be referred to for explanation and will not be described in detail.

[0181] Figure 2 FIG. 1 is a schematic diagram of a detection device for high-voltage GIS equipment provided by an embodiment of the present invention. Figure 2 As shown, the detection device of high-voltage GIS equipment provided by the embodiment of the present invention includes an acquisition unit 201, a determination unit 202 and a detection unit 203, wherein:

[0182] The acquisition unit 201 is used to acquire the substation environment image including the high-voltage GIS equipment, and the three-dimensional point cloud data after registration of the substation environment; the determination unit 202 is used to determine the candidate target area for detecting the high-voltage GIS equipment based on the substation environment image and the three-dimensional point cloud data; the candidate target area includes the key parts of the high-voltage GIS equipment; the detection unit 203 is used to verify the candidate target area according to the pre-established three-dimensional model of the GIS equipment, obtain the target area position that meets the equipment characteristics, and control the actuator to detect the target area position.

[0183] Specifically, the acquisition unit 201 in the device is used to acquire an environmental image of a substation including a high-voltage GIS device, and three-dimensional point cloud data after registration of the substation environment; the determination unit 202 is used to determine a candidate target area for detecting the high-voltage GIS device based on the substation environmental image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS device; the detection unit 203 is used to verify the candidate target area based on a pre-established three-dimensional model of the GIS device, obtain the target area position that meets the equipment characteristics, and control the actuator to detect the target area position.

[0184] The detection device for high-voltage GIS equipment provided in the embodiment of the present invention obtains an environmental image of a substation including the high-voltage GIS equipment, and three-dimensional point cloud data after registration of the substation environment; determines a candidate target area for detecting the high-voltage GIS equipment based on the substation environmental image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS equipment; verifies the candidate target area based on a pre-established three-dimensional model of the GIS equipment, obtains the target area position that meets the equipment characteristics, and controls the actuator to detect the target area position, which can improve the safety and efficiency of high-voltage GIS equipment detection.

[0185] The embodiment of the present invention provides an embodiment of a detection device for high-voltage GIS equipment, which can be specifically used to execute the processing flow of the above-mentioned method embodiments. Its functions are not repeated here, and reference can be made to the detailed description of the above-mentioned method embodiments.

[0186] Figure 3 A schematic diagram of the physical structure of a computer device provided in an embodiment of the present invention is shown in FIG. Figure 3 As shown, the computer device includes: a memory 301, a processor 302, and a computer program stored in the memory 301 and executable on the processor 302. When the processor 302 executes the computer program, the following method is implemented:

[0187] Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment;

[0188] Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment;

[0189] The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

[0190] This embodiment discloses a computer program product, the computer program product including a computer program, and when the computer program is executed by a processor, the following method is implemented:

[0191] Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment;

[0192] Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment;

[0193] The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

[0194] This embodiment provides a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the following method is implemented:

[0195] Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment;

[0196] Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment;

[0197] The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

[0198] Compared with the technical solutions in the prior art, the detection method of high-voltage GIS equipment provided in the embodiment of the present invention obtains a substation environment image containing the high-voltage GIS equipment, and three-dimensional point cloud data after registration of the substation environment; determines a candidate target area for detecting the high-voltage GIS equipment based on the substation environment image and the three-dimensional point cloud data; the candidate target area includes key parts of the high-voltage GIS equipment; verifies the candidate target area based on a pre-established three-dimensional model of the GIS equipment, obtains the target area position that meets the equipment characteristics, and controls the actuator to detect the target area position, which can improve the safety and efficiency of high-voltage GIS equipment detection.

[0199] Those skilled in the art will appreciate that embodiments of the present invention may be provided as methods, systems, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware. Moreover, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0200] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0201] These computer program instructions may also be stored in a computer-readable memory capable of directing a computer or other programmable data processing device to operate in a specific manner, so that the instructions stored in the computer-readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.

[0202] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operating steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing instructions for implementing the process in the computer or other programmable device. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.

[0203] In the description of this specification, the description with reference to the terms "one embodiment", "a specific embodiment", "some embodiments", "for example", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present invention. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.

[0204] The specific embodiments described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A detection method for high-voltage GIS equipment, characterized in that: include: Obtain the substation environment image including the high-voltage GIS equipment, and the registered 3D point cloud data of the substation environment; Determine a candidate target area for detecting the high-voltage GIS equipment according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS equipment; The candidate target area is verified according to the pre-established three-dimensional model of the GIS equipment, the position of the target area that meets the equipment characteristics is obtained, and the control actuator is controlled to detect the position of the target area.

2. The detection method of high-voltage GIS equipment according to claim 1, characterized in that: Acquiring the three-dimensional point cloud data includes: Preprocess the initial three-dimensional point cloud data of the power station environment obtained by LiDAR; Insert the preprocessed initial 3D point cloud data into the local map, and mark the 3D point cloud data inserted into the local map as active; Perform scene registration on the three-dimensional point cloud data in an active state to obtain the three-dimensional point cloud data.

3. The detection method of high-voltage GIS equipment according to claim 2, characterized in that: The performing scene registration on the active three-dimensional point cloud data to obtain the three-dimensional point cloud data includes: Based on the SC2-PCR algorithm, the directional similarity measurement is calculated for the active three-dimensional point cloud data with preliminary correspondence; The three-dimensional point cloud data whose direction similarity measurement calculation results are higher than a threshold value are determined as a first seed point set with high reliability, and the remaining three-dimensional point cloud data are determined as a second seed point set with low reliability; Collectively sampling the first seed point set and the second seed point set according to a preset ratio, and performing an overall consistency evaluation to obtain an optimal transformation matrix; The three-dimensional point cloud data in the active state is registered and calculated according to the optimal transformation matrix to obtain the three-dimensional point cloud data.

4. The detection method of high-voltage GIS equipment according to claim 1, characterized in that: Acquiring the substation environment image, including: Performing initial preprocessing on the initial substation environment image including high-voltage GIS equipment obtained by the depth camera; The initial substation environment image after the primary preprocessing is used as the substation environment image.

5. The detection method of high-voltage GIS equipment according to claim 1, characterized in that: The determining of a candidate target area for detecting the high-voltage GIS device according to the substation environment image and the three-dimensional point cloud data includes: Performing secondary preprocessing on the substation environment image to obtain an optimized substation environment image; Acquire edge features of the high-voltage GIS equipment according to the optimized substation environment image; Performing secondary preprocessing on the three-dimensional point cloud data to obtain optimized three-dimensional point cloud data; Acquire key geometric features of the high-voltage GIS equipment according to the optimized three-dimensional point cloud data; The edge features and the key geometric features are fused, and the candidate target area is determined according to the feature fusion result.

6. The detection method of high-voltage GIS equipment according to claim 2, characterized in that: The control execution mechanism detects the position of the target area, including: Sending a control instruction to a mechanical arm mounted on the UAV so that the mechanical arm detects the position of the target area according to the control instruction; Wherein, the control instruction carries the target area position information corresponding to the target area position; the drone is also equipped with a laser radar and a depth camera.

7. The detection method of high-voltage GIS equipment according to claim 6, characterized in that: The detection method of the high-voltage GIS equipment also includes: If it is determined that the positioning of the UAV fails during the process of inserting the pre-processed initial three-dimensional point cloud data into the local map, the UWB device is used to reposition the UAV.

8. A detection device for high-voltage GIS equipment, characterized in that: include: An acquisition unit, used to acquire an environmental image of a substation including high-voltage GIS equipment, and registered three-dimensional point cloud data of the substation environment; A determination unit, configured to determine a candidate target area for detecting the high-voltage GIS device according to the substation environment image and the three-dimensional point cloud data; the candidate target area includes a key part of the high-voltage GIS device; The detection unit is used to verify the candidate target area according to the pre-established three-dimensional model of the GIS equipment, obtain the target area position that meets the equipment characteristics, and control the actuator to detect the target area position.

9. A computer device comprising a memory, a processor and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 7 is implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images