An overhaul method and system for a rotating machine based on digital twin technology

Abstract

The present invention relates to the technical field of test and maintenance, and particularly to an overhaul method and system for rotating equipment based on digital twin technology. The method includes constructing a three-dimensional virtual model of a substation including a physical space, a virtual space, and a digital service space; obtaining sensor data of an object to be modeled through a sensor component, processing the data, constructing a three-dimensional model and filling it into the virtual space; mapping the physical space and the virtual space to the digital service space, and in the digital service space, the operation status of the equipment, fault location information are displayed in real time, and overhaul suggestions are provided. An overhaul device is controlled by generating an overhaul instruction to overhaul the rotating equipment, so as to realize efficient monitoring and maintenance of substation equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of test and maintenance, and particularly to a maintenance method and system for rotating equipment based on digital twin technology. Background Art

[0002] With the increasing complexity of the power system and the improvement of the automation level, the substation, as a key node for power transmission and distribution, has gradually become an important part of the power grid operation. To ensure the safe and stable operation of the power system, it is necessary to monitor various types of equipment in the substation, especially the health status and operation efficiency of rotating equipment. However, the traditional maintenance methods for substation equipment mainly rely on regular maintenance and passive fault detection. This method not only has low efficiency but also has problems such as insufficient detection accuracy, inaccurate fault location, and high maintenance costs.

[0003] In the prior art, digital twin technology is introduced to monitor the power system. Digital twin technology realizes real-time interaction and mapping between the physical space and the virtual space by constructing a digital virtual model of physical equipment, providing technical support for equipment operation monitoring, fault diagnosis, maintenance decision-making, etc. Digital twin technology can not only monitor the operation status of equipment in real time but also perform fault prediction and maintenance optimization through simulation, thus significantly improving the efficiency and reliability of equipment management.

[0004] For example, the patent with the publication number CN107168238A discloses an integrated system for substation equipment maintenance and information management, including a controller, which is communicatively connected to a touch panel and a data memory. It also includes a two-dimensional code identifier for scanning and identifying two-dimensional codes; the data memory is used to store relevant information of the equipment, documents recording equipment operation methods and operation steps, and operation videos; the two-dimensional code is used to represent relevant information of the equipment; the touch panel is used to display documents recording equipment operation methods and operation steps and operation videos; the user scans the two-dimensional code through the two-dimensional code identifier to obtain two-dimensional code information, and obtains relevant information of the represented equipment through the two-dimensional code information; when the controller obtains the two-dimensional code information through the user, the relevant information of the represented equipment is matched with the relevant information of the equipment in the data memory; when the matching is passed, the touch panel is triggered to display information. This invention solves the problems of scattered and non-centralized existing equipment data information and low maintenance efficiency.

[0005] The above patent has the problems proposed in this background art: The rotating equipment in the substation involves different sensor data. How to effectively integrate and process this heterogeneous data to generate a high-precision three-dimensional model is still an unsolved problem. To solve the above problems, this application designs a maintenance method and system for rotating equipment based on digital twin technology. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a maintenance method and system for rotating equipment based on digital twin technology in view of the deficiencies of the prior art. The method includes constructing a three-dimensional virtual model of a substation including a physical space, a virtual space, and a digital service space; obtaining sensor data of an object to be modeled through a sensor component, processing the data, constructing a three-dimensional model and filling it into the virtual space; mapping the physical space and the virtual space to the digital service space, and displaying the equipment operation status, fault location information, and providing maintenance suggestions in real time through the digital service space, and controlling a maintenance device to perform maintenance on the rotating equipment by generating a maintenance instruction.

[0007] To achieve the above object, the present invention provides the following technical solutions:

[0008] A maintenance system for rotating equipment based on digital twin technology, the system includes an equipment virtual module, an equipment operation status monitoring module, and a test and maintenance control module;

[0009] The equipment virtual module is used for three-dimensional visualization of the rotating equipment in the substation, and the equipment virtual module is configured with a data processing strategy and a three-dimensional modeling strategy;

[0010] The equipment operation status monitoring module is used for monitoring the rotating equipment through a visualization interface, and performing real-time analysis on the monitoring data, identifying abnormal data patterns, and calculating the analysis results;

[0011] The test and maintenance control module is used for generating a test and maintenance plan according to the analysis results, converting the test and maintenance plan into a machine control instruction, and outputting the machine control instruction;

[0012] The test and maintenance control module is configured with a maintenance control strategy, and the maintenance control strategy analyzes the analysis results according to an intelligent decision-making algorithm, generates a test and maintenance plan, and generates a maintenance instruction through a control algorithm, and controls a maintenance device to perform maintenance on the rotating equipment through the maintenance instruction.

[0013] The equipment operation status monitoring module includes:

[0014] An equipment monitoring unit: used for monitoring the rotating equipment through a sensor to obtain the monitoring data of the equipment;

[0015] A real-time data analysis unit: performing real-time analysis on all monitoring data, identifying abnormal data patterns and calculating the analysis results;

[0016] The equipment operation status monitoring module is configured with a data analysis strategy, and the data analysis strategy includes a temperature analysis logic, a vibration analysis logic, a pressure analysis logic, a current analysis logic, and a flow analysis logic;

[0017] The temperature analysis logic is used to analyze the temperature data collected by the temperature sensor to determine whether there is an overheating fault;

[0018] The pressure analysis logic is used to analyze the pressure data collected by the pressure sensor to identify the trend and fluctuation of the pressure over time and determine whether there is overpressure and pressure relief;

[0019] The flow analysis logic is used to analyze the flow data collected by the turbine flowmeter to determine whether there is a situation of over-limit flow;

[0020] The temperature analysis logic, vibration analysis logic, pressure analysis logic, current analysis logic and flow analysis logic are configured in the real-time data analysis unit.

[0021] The vibration analysis logic is used to analyze the vibration signal collected by the vibration sensor to determine whether there is abnormal vibration of the equipment bearing, specifically including:

[0022] Perform signal reconstruction on the vibration signal;

[0023] Perform Fourier transform on the vibration signal after signal reconstruction, and divide the Fourier spectrum of the vibration signal into frequency bands with different center frequencies;

[0024] Iterate for each frequency band to calculate the optimal solution under the condition that the sum of the bandwidths satisfying each modal component is the smallest;

[0025] Calculate the power spectrum according to the optimal solution of each modal component, obtain the characteristic frequency of the rotating equipment bearing signal, compare the characteristic frequency of the rotating equipment bearing signal with the theoretical frequency of the rotating equipment bearing signal, and output the vibration signal analysis result.

[0026] The current analysis logic is used to analyze the current signal collected by the current sensor to determine whether there is load startup and current overload, specifically including:

[0027] Rectify and smooth the current signal, perform frequency domain analysis on the current signal through short-time Fourier transform, and extract high-frequency components and low-frequency components;

[0028] Compare the high-frequency component of the current signal with the current detection threshold. If it is greater than or equal to the current detection threshold, it indicates that there is current overload;

[0029] Compare the low-frequency component of the current signal with the initial current threshold. If it is greater than or equal to the initial current threshold, it indicates that there is load startup.

[0030] The test and maintenance control module includes:

[0031] A solution generation unit, configured to process the analysis results and generate a test and maintenance solution with the minimum cost;

[0032] A solution execution unit, configured to convert the test and maintenance solution into machine instructions executable by the system, and control a maintenance device to perform maintenance on the rotating equipment through the machine instructions;

[0033] An execution feedback unit, configured to monitor the operating status of the rotating equipment in real time, compare it with the expected maintenance results, and automatically adjust the maintenance control strategy and machine control instructions according to the results of the real-time feedback;

[0034] The maintenance control strategy includes an optimal solution generation logic and a control logic. The optimal solution generation logic is configured in the solution generation unit, and the control logic is configured in the solution execution unit.

[0035] The optimal solution generation logic includes:

[0036] Construct a genetic algorithm, and define the objective function and constraints;

[0037] Use the analysis results and the equipment operating status as the input parameters of the genetic algorithm, perform iterative optimization on the input parameters through the genetic algorithm, and output the optimal solution of the objective function;

[0038] Use the optimal solution of the objective function as the test and maintenance solution.

[0039] The equipment virtual module includes:

[0040] A three-dimensional virtual model construction unit, configured to construct a three-dimensional virtual model of the substation, and perform data mapping and interaction between the physical space, the virtual space, and the digital service space;

[0041] A sensor data acquisition unit, configured to obtain the external image of the object to be modeled and the laser scanning point cloud of the object to be modeled through a variety of sensor components;

[0042] A data processing unit, configured to process the external data and internal data of the object to be modeled, and perform three-dimensional visualization on the rotating equipment in the substation.

[0043] The three-dimensional virtual model construction unit includes:

[0044] The physical space unit: includes physical devices and their layouts, and collects physical data of the devices through sensors;

[0045] The virtual space unit: digitally represents the devices and environment in the physical space, and includes the three-dimensional models of the devices, the virtual environment, and the simulation data;

[0046] Digital service space unit: a digital service for providing real-time monitoring, data analysis, and maintenance guidance;

[0047] The data processing strategy includes an image processing logic and a point cloud processing logic. The image processing logic is used to extract the surface features of the rotating equipment through image processing techniques, and the point cloud processing logic is used to extract the internal features of the rotating equipment;

[0048] The data processing unit includes:

[0049] Image processing sub-unit: processes external data and extracts external structural features;

[0050] Point cloud processing sub-unit: processes internal data and extracts internal point cloud features;

[0051] Feature fusion sub-unit: fuses the external structural features and the internal point cloud features and generates a three-dimensional model of the object to be modeled;

[0052] Model merging sub-unit: merges multiple three-dimensional models to construct a complete three-dimensional model of the substation;

[0053] The three-dimensional modeling strategy includes a feature fusion logic, which is used to fuse the external structural features and the internal point cloud features.

[0054] The image processing logic is configured in the image processing sub-unit, and the point cloud processing logic is configured in the point cloud processing sub-unit;

[0055] The feature fusion logic is configured in the feature fusion sub-unit.

[0056] The image processing logic includes:

[0057] Compresses the external image of the object to be modeled to obtain a compressed reconstructed image;

[0058] Calibrates the image sensor to obtain the external parameter matrix and the internal parameter matrix of the image sensor, performs coordinate transformation on the compressed reconstructed image according to the external parameter matrix and the internal parameter matrix, and converts the pixels of the compressed reconstructed image from the camera coordinate system to the world coordinate system;

[0059] Calculates the global mean of all pixel points of the compressed reconstructed image after coordinate transformation, performs weighted sampling through an upsampling function, and calculates the global feature map;

[0060] Compares the confidence of the pixel points of the global feature image with the confidence threshold, retains the pixel points greater than the confidence threshold, and obtains the surface feature map;

[0061] Perform regional cutting on the surface feature map, extract the regional time features of the cut surface feature map according to the time feature extraction function, and extract the regional spatial features of the cut surface feature map through the attention mechanism;

[0062] Fuse the regional time features and regional spatial features of the surface feature map to obtain bilinear regional features, and calculate the external structure features by pooling the bilinear regional features of all regions through the pooling function.

[0063] The calculation formula of the external structure feature is:

[0064]

[0065] Among them, Gb fd represents the external structure feature, a represents the cutting region of a single surface feature map, A represents the total number of cutting regions of the surface feature map, b{·} represents the pooling function, T f (·) represents the time feature extraction function, hb fi represents the surface feature map, * represents the fusion operation, Ex[·] represents the excitation function in the attention mechanism, W represents the horizontal length of the surface feature map, H represents the vertical length of the surface feature map, n represents the unit horizontal length of the surface feature map, m represents the unit vertical length of the surface feature map, v a represents the convolution kernel corresponding to the cutting region.

[0066] The point cloud processing logic includes:

[0067] Preprocess the laser scan point cloud according to the laser scan point cloud of the internal components of the object to be modeled;

[0068] Find the most suitable three-dimensional plane through plane fitting, perform coordinate transformation on the preprocessed three-dimensional point cloud data, and transfer the laser scan point cloud from the radar coordinate system to the world coordinate system;

[0069] Divide the world coordinate system into grids at equal intervals, represent the laser scan point cloud in voxel form, project the laser scan point cloud according to the voxel index into the corresponding grids, connect the points within the voxels to strengthen the local feature information, and obtain the voxel features of the laser scan point cloud;

[0070] Perform clustering analysis on the voxel features of the laser scan point cloud according to the clustering algorithm, merge the grids where the voxel features of the laser scan point cloud under the same central cluster are located, and extract the internal point cloud features. The calculation formula of the internal point cloud feature is:

[0071]

[0072] Among them, T aDenote the internal point cloud features corresponding to the voxel features of the laser scan point cloud that are in the same central cluster after clustering. Let \(i\) represent a single voxel feature of the laser scan point cloud in the central cluster, \(M\) represent the total number of voxel features of the laser scan point cloud in the central cluster, and \(\eta\). i Denote the eigenvalue of the \(i\)-th voxel feature of the laser scan point cloud, \(d\) denote the clustering threshold corresponding to the central cluster, and \(r\). 2 Denote the farthest distance scanned by the laser scanner, and \(r\). 1 Denote the nearest distance scanned by the laser scanner, and \(p\). i_x Denote the voxel abscissa of the \(i\)-th voxel feature of the laser scan point cloud, and \(p\). i_y Denote the voxel ordinate of the \(i\)-th voxel feature of the laser scan point cloud, and \(p\). i_z Denote the voxel vertical coordinate of the \(i\)-th voxel feature of the laser scan point cloud.

[0073] The clustering analysis includes:

[0074] Perform preliminary clustering on the voxel features of the laser scan point cloud through the DBSCAN clustering algorithm to obtain core points;

[0075] Take the core points after DBSCAN clustering as the initial central clusters of the K-means clustering algorithm, and calculate the Euclidean distances from the voxel features of the laser scan point cloud to their corresponding initial central clusters;

[0076] Perform clustering division according to the Euclidean distances, calculate the clustering average error of the initial central clusters, update the central clusters according to the clustering average error, and determine whether the updated central clusters meet the convergence condition. If they do not meet the convergence condition, continue to update the central clusters. If they meet the convergence condition, output the clustering result.

[0077] The feature fusion logic includes:

[0078] Convert the external structure features and the internal point cloud features into feature sequences, where the feature sequences include image point cloud feature sequences and laser scan point cloud feature sequences;

[0079] Construct a geometric feature extraction module, a surface feature extraction module, an internal feature extraction module, and a spatial feature extraction module;

[0080] Process the image point cloud feature sequence through the geometric feature extraction module and the surface feature extraction module to extract geometric features and surface features;

[0081] Process the laser scan point cloud feature sequence through the internal feature extraction module and the spatial feature extraction module to extract internal features and spatial features;

[0082] Map the geometric features, the surface features, the internal features, and the spatial features to tuple orders, and calculate the fused features through pooling operations and residual connections.

[0083] The surface feature extraction module is used to extract and integrate surface shape features and surface structure features from the image point cloud feature sequence. The surface feature extraction module includes a surface shape feature extraction sub-module, a surface structure feature extraction sub-module, and a feature concatenation sub-module;

[0084] The geometric feature extraction module is used to identify and extract geometric shape features from the image point cloud feature sequence. The geometric feature extraction module includes a geometric shape detection sub-module, a normal vector calculation sub-module, and a curvature estimation sub-module;

[0085] The spatial feature extraction module is used to extract the spatial layout features of an object from the laser scan point cloud feature sequence. The spatial feature extraction module includes a spatial encoding sub-module, a spatial transformation sub-module, and a spatial feature aggregation sub-module;

[0086] The internal feature extraction module is used to extract the internal material distribution features of an object from the laser scan point cloud feature sequence.

[0087] The calculation formula for the fused feature is:

[0088] F fu = b{Con[En(T 1 ), En(T 2 ),..., En(T r )]},

[0089] where F fu represents the fused feature, Con[·] represents the residual connection function, En(·) represents the local enhancement function, and T r represents the r-th feature group in the tuple order.

[0090] A maintenance method for rotating equipment based on digital twin technology, the method includes:

[0091] Construct a three-dimensional virtual model of a substation, and the three-dimensional virtual model of the substation includes a physical space, a virtual space, and a digital service space;

[0092] According to the physical space, determine the object to be modeled, and obtain the sensor data of the object to be modeled through the sensor component;

[0093] Process the sensor data, construct a three-dimensional model of the substation, and fill the three-dimensional model of the substation into the virtual space;

[0094] Map the physical space and the virtual space to the digital service space, and display the operating status, fault location information, and test and maintenance plans of the device in real time through the digital service space;

[0095] Generate a maintenance instruction according to the test and detection plan, and control the maintenance device to perform maintenance on the rotating equipment through the maintenance instruction.

[0096] Compared with the prior art, the beneficial effects of the present invention are:

[0097] 1. By constructing a three-dimensional virtual model including a physical space, a virtual space, and a digital service space, the present invention realizes a high degree of synchronization between the physical space and the virtual space. The present invention can obtain device data in the physical space in real time through a sensor component, and integrate these data into the virtual space to generate an accurate three-dimensional model. This three-dimensional model can not only display the geometric shape of the device, but also reflect the operating status of the device, providing comprehensive visual support for device maintenance and fault troubleshooting.

[0098] 2. The present invention obtains external images and point cloud data through an image sensor and a laser scanner, extracts external structure features and internal point cloud features, generates a three-dimensional model through feature fusion and modeling software, and finally performs fault troubleshooting and early warning to realize efficient monitoring and maintenance of the rotating equipment in the substation. BRIEF DESCRIPTION OF THE DRAWINGS

[0099] By reading the detailed description of the non-limiting embodiments with reference to the following drawings, other features, objects, and advantages of the present invention will become more apparent:

[0100] Figure 1 It is a schematic flow chart of a method for overhauling rotating equipment based on digital twin technology in Embodiment 1 of the present invention;

[0101] Figure 2 It is a schematic structural diagram of a three-dimensional virtual model of a substation in Embodiment 1 of the present invention;

[0102] Figure 3 It is a flow chart for extracting external structure features disclosed by the present invention;

[0103] Figure 4 It is a flow chart for clustering voxel features of laser scanning point clouds disclosed by the present invention;

[0104] Figure 5 It is a diagram of the grid clustering result disclosed by the present invention;

[0105] Figure 6 It is a structural diagram of a feature fusion network disclosed by the present invention;

[0106] Figure 7 It is a structural diagram of a surface feature extraction module disclosed by the present invention;

[0107] Figure 8 Structural diagram of the geometric feature extraction module disclosed by the present invention;

[0108] Figure 9 Structural diagram of the spatial feature extraction module disclosed by the present invention;

[0109] Figure 10 Schematic structural diagram of an overhaul system for rotating equipment based on digital twin technology according to the present invention. Specific implementation manners

[0110] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments.

[0111] Embodiment 1

[0112] Please refer to Figure 1 , an embodiment provided by the present invention: a method for overhauling rotating equipment based on digital twin technology, and the specific steps of the method are as follows:

[0113] S1: Construct a three-dimensional virtual model of a substation, determine the object to be modeled, and the three-dimensional virtual model of the substation includes a physical space, a virtual space, and a digital service space;

[0114] S2: Obtain the external image of the object to be modeled through an image sensor, process the external image, and extract the external structural features of the object to be modeled;

[0115] S3: Obtain the laser scanning point cloud of the object to be modeled through a laser scanner, process the laser scanning point cloud, and extract the internal point cloud features of the object to be modeled;

[0116] S4: Construct a feature fusion network, use the external structural features and the internal point cloud features as input parameters of the feature fusion network, train the input parameters through the feature fusion network, and calculate the fusion features;

[0117] S5: Import the fusion features into modeling software for processing, output the three-dimensional model of the object to be modeled, merge all the three-dimensional models, construct a three-dimensional model of the substation, and fill the three-dimensional model of the substation into the virtual space;

[0118] S6: Map the physical space and the virtual space to the digital service space, and monitor the operating state of the rotating equipment of the substation through the digital service space;

[0119] S7: Based on the monitoring results, conduct troubleshooting on the suspected faulty equipment, locate the numbers and positions of the faulty equipment, display them in the digital service space, and formulate a maintenance plan.

[0120] S8: Convert the maintenance plan into a maintenance instruction, and control the maintenance device to perform maintenance on the rotating equipment through the maintenance instruction.

[0121] Please refer to Figure 2 , the schematic diagram of the three-dimensional virtual model structure of the substation in the embodiment of the present invention. The physical space includes the actual physical equipment in the substation, and the virtual space is the digital representation of the physical space, including virtual objects and virtual environments. The virtual space is a digital space constructed from different granularities based on computer technology according to application needs, that is, the digital mapping of the substation physical equipment in the computer environment, and describes the geometric structure characteristics and external behavioral manifestations of the physical equipment object through the computer. Therefore, the virtual entity not only includes the three-dimensional geometric models of entity equipment objects, such as the three-dimensional model of the transformer, the three-dimensional model of the high-voltage switchgear, the three-dimensional model of the busbar equipment, the three-dimensional model of the capacitor, the three-dimensional model of the distribution device, and the three-dimensional model of the cable equipment, but also includes the geometric constraint model, the dynamic function behavior model, and the fault behavior model. The digital service space is used to provide digital services such as real-time monitoring, data analysis, fault diagnosis, and maintenance guidance.

[0122] The physical space can be regarded as a perceivable physical entity, which is the target object of maintenance digitization. Mastering the physical space is the prerequisite for realizing three-dimensional digitization. The physical space includes physical objects and sensor components. The physical objects include transformers, high-voltage switchgear, busbar equipment, capacitors, distribution devices, and cable equipment. The sensor components include image sensors, laser scanners, and monitoring sensors. The transformer includes the main transformer and the auxiliary transformer. The main transformer includes the outer shell, the cooling system, the radiator, and the insulating oil tank. The auxiliary transformer is an auxiliary power transformer for the power supply system. The high-voltage switchgear includes circuit breakers, disconnectors, and load switches. The circuit breaker is used to cut off or close the circuit, and its outer shell, operating mechanism, and internal contacts need to be modeled. The disconnector is used for circuit isolation, and its knife switch, transmission mechanism, and other components should be modeled. The load switch is used to carry and cut off the load current, and its outer shell and operating mechanism need to be modeled. The busbar equipment is a conductor for transmitting high-voltage current, and its busbar support insulators, connection clamps, and fixing devices need to be modeled. The capacitor includes a capacitor bank and a reactor. The capacitor bank is used for reactive power compensation, and its capacitors, connecting wires, and insulating supports should be modeled. The reactor is used for current limiting or filtering, and its reactor coil and iron core need to be modeled. The distribution device is a device for distributing electric power, and its distribution device outer shell, internal switches, protection devices, and terminal blocks need to be modeled. The cable equipment includes high-voltage and low-voltage cables.

[0123] In the process of constructing a digital twin, the physical space serves as the object for data collection. Through devices such as image sensors and laser scanners, the external and internal characteristics of the devices in the physical space are obtained. These characteristic data are the basis for subsequent modeling and analysis. The accurate description of the physical space helps to ensure the accuracy of the data and provides reliable data support for the construction of the virtual space.

[0124] Please refer to Figure 3 , the flowchart of external structure feature extraction in the embodiment of the present invention. The specific steps of S2 are as follows:

[0125] S2.1: Preprocess the external image of the object to be modeled. The preprocessing includes image smoothing and edge enhancement. Compress the preprocessed image through an image compression algorithm to generate a compressed reconstructed image;

[0126] Reduce noise and enhance the contrast of the image through smoothing and edge enhancement. The compression algorithm should ensure that while reducing the size of the image file, as much key structural information in the image as possible is retained;

[0127] S2.2: Calibrate the image sensor to obtain the external parameter matrix and internal parameter matrix of the image sensor. Camera calibration is a technique for obtaining the internal and external parameters of a camera through multiple images at different angles. The internal parameter matrix describes the internal attributes of the camera, such as focal length, position of the optical center, etc. The external parameter matrix describes the position and orientation of the camera in the world coordinate system. Perform coordinate transformation on the compressed reconstructed image according to the external parameter matrix and internal parameter matrix, and transfer the pixels of the compressed reconstructed image from the camera coordinate system to the world coordinate system. Through coordinate transformation, associate the two-dimensional pixel positions in the image with the three-dimensional coordinates in the real physical space to provide more accurate spatial positioning;

[0128] S2.3: Calculate the global mean of all pixel points of the compressed reconstructed image after coordinate transformation, extract the local feature map, perform weighted sampling on the local feature map through an upsampling function, superimpose the weighted sampled local feature map on the compressed reconstructed image, calculate the global feature map, and compare the confidence of the global feature map pixels with the confidence threshold, retain the pixel points greater than the confidence threshold, and obtain the surface feature map to enhance the characterization ability of the image features. The global mean calculation captures the overall features of the image, while the local feature extraction focuses on the details in the image. By superimposing the compressed reconstructed image and the local feature map, the importance of the local features can be enhanced globally, thus providing a more comprehensive feature representation for subsequent feature fusion;

[0129] S2.4: Perform regional cutting on the surface feature map, dividing the feature map into multiple small regions, where each small region represents different structural parts in the image. Then, use the time feature extraction function to extract the time-dimensional features of each region to obtain the change information at different time points. Next, use the attention mechanism to extract the spatial features of each region. The attention mechanism can focus on the key regions in the image and extract representative spatial information. The regional cutting operation makes the feature extraction more refined and can capture the features of different parts of the image. The time feature extraction function provides the dynamic information of each region, while the attention mechanism ensures the accurate extraction of spatial features. Through this multi-dimensional feature extraction, the model can better understand the complex structure and dynamic changes of the image;

[0130] S2.5: After obtaining the time features and spatial features of each region, use the bilinear pooling operation to fuse these features. Bilinear pooling realizes the full combination of time features and spatial features by calculating the outer product of the two features, generating bilinear region features. Then, use the pooling function to comprehensively process the bilinear region features of all regions to generate external structure features. The feature fusion process fully integrates the feature information of different dimensions, enabling the external structure features to comprehensively reflect the structural details and overall morphology in the image. The pooling operation further refines these features, removes redundant information, and enhances the discriminative ability of the model. The calculation formula for the external structure features is as follows:

[0131]

[0132] where, Gb fd represents the external structure features, a represents the cutting region of a single surface feature map, A represents the total number of cutting regions of the surface feature map, b{·} represents the pooling function, T f (·) represents the time feature extraction function, hb fi represents the surface feature map, * represents the fusion operation, Ex[·] represents the excitation function in the attention mechanism, W represents the horizontal length of the surface feature map, H represents the vertical length of the surface feature map, n represents the unit horizontal length of the surface feature map, m represents the unit vertical length of the surface feature map, and v a represents the convolution kernel corresponding to the cutting region.

[0133] The main purpose of the compression process is to reduce the redundant information in the image, lower the overhead of data storage and processing, and at the same time maintain the integrity of the key feature information. This step effectively improves the efficiency of subsequent image processing and provides efficient input data for feature extraction and 3D modeling. The specific steps of S2.1 are as follows:

[0134] S2.1.1: Denoise the surface image, where the noise includes on-site light noise, the operating noise of each unit itself, circuit switch noise, and the photoelectric tube noise of the image sensor;

[0135] S2.1.2: Perform sparse representation on the denoised surface image, and perform masked sampling through Fourier transform to obtain a sparse surface image;

[0136] S2.1.3: Cut the sparse surface image, cut the original image with size h*w into blocks with size p*q to obtain a block image matrix, construct a sequence of observation matrices, match different sampling rates according to the composition complexity of different regions of the block image, and perform sparse sampling compression in the Fourier domain in a distributed multi-point cooperation manner;

[0137] S2.1.4: Divide the preprocessed image into segmented images with the row length = column length = γ according to matrix binarization, obtain the centroid position of the segmented image, calculate the box dimension of the segmented image according to the centroid position, and compare the box dimension of the segmented image with the background threshold;

[0138] S2.1.5: If it is greater than or equal to the background threshold, retain the segmented image. If it is less than the background threshold, delete the segmented image. Traverse all segmented images and fuse the remaining segmented images to obtain a compressed reconstructed image, where γ represents the row and column lengths.

[0139] The specific steps of S3 are as follows:

[0140] S3.1: Preprocess the laser scan point cloud according to the laser scan point cloud of the internal components of the object to be modeled;

[0141] S3.2: During the point cloud data acquisition process, there are differences in the pose and position of the object to be modeled in scans at different angles. According to the least squares method, by adjusting the direction and position of the plane, the sum of the squares of the distances from all points to the plane is minimized, and the most suitable three-dimensional plane is found through plane fitting. Perform coordinate transformation on the preprocessed three-dimensional point cloud data, transfer the laser scan point cloud from the radar coordinate system to the world coordinate system, and unify the surface point cloud data in the same coordinate system under scans at different angles, providing spatial consistency for subsequent data processing and feature extraction;

[0142] S3.3: Divide the world coordinate system at equal intervals into grids, voxelize the laser scan point cloud, project it into the corresponding grids according to the voxel indices of the laser scan point cloud, connect the points within the voxels to strengthen the local feature information, and obtain the voxel features of the laser scan point cloud. The voxelization process divides the space into uniform small cube units and aggregates the point sets located within the same voxel into a voxel point. This representation method not only reduces the data volume but also better preserves and represents the detailed features of the object to be modeled by strengthening the local feature information.

[0143] S3.4: Perform preliminary clustering on the voxel features of the laser scan point cloud according to the DBSCAN clustering algorithm to obtain core points, and determine the number of cluster centers based on the number of the core points. DBSCAN classifies the points in the high-density area into the same class by analyzing the density of the points, and can effectively identify and ignore the noise points at the same time. Determine the preliminary number of cluster centers based on the number of core points. The clustering result of this step helps to identify the main areas of the internal structure of the object to be modeled and lays a foundation for the subsequent precise clustering.

[0144] S3.5: Use the core points after DBSCAN clustering as the initial cluster centers of the K-means clustering algorithm, calculate the Euclidean distances from the voxel features of the laser scan point cloud to each of the initial cluster centers. Use the core points obtained by DBSCAN clustering as the initial cluster centers of the K-means algorithm to further optimize the clustering result by the K-means algorithm. K-means performs clustering division by calculating the Euclidean distances from the voxel features to the initial cluster centers. The selection of the initial cluster centers has a great influence on the final convergence effect of the K-means algorithm. By using the DBSCAN clustering result as the initial points, the convergence speed and clustering quality of the K-means algorithm can be improved.

[0145] S3.6: Perform clustering division according to the Euclidean distances, calculate the clustering average error of the initial cluster centers, update the cluster centers according to the clustering average error, and judge whether the updated cluster centers meet the convergence condition. If they do not meet the convergence condition, continue to update the cluster centers. If they meet the convergence condition, output the clustering result. In each iteration, the K-means algorithm reassigns each point to the nearest cluster center and updates the positions of the cluster centers. This process continues to iterate until the average error of the clustering meets the convergence condition and the clustering result is stable. Through this process, the algorithm can more accurately extract the internal features of the object to be modeled and finally output the clustering result.

[0146] S3.7: Extract the voxel features of the clustered laser scan point cloud. Merge the grids where the voxel features of the laser scan point cloud under the same central cluster are located, ensuring that the feature points belonging to the same class are aggregated together, making the classification of voxel features more accurate. The finally extracted point cloud features can reflect the internal morphological features and potential defects. Extract the internal point cloud features. The calculation formula for the internal point cloud features is as follows:

[0147]

[0148] Among them, T a represents the internal point cloud features corresponding to the voxel features of the laser scan point cloud in the same central cluster after clustering. i represents a single voxel feature of the laser scan point cloud in the central cluster. M represents the total number of voxel features of the laser scan point cloud in the central cluster. η i represents the eigenvalue of the i-th voxel feature of the laser scan point cloud. d represents the clustering threshold corresponding to the central cluster. r 2 represents the farthest distance scanned by the laser scanner. r 1 represents the nearest distance scanned by the laser scanner. p i_x represents the voxel abscissa of the i-th voxel feature of the laser scan point cloud. p i_y represents the voxel ordinate of the i-th voxel feature of the laser scan point cloud. p i_z represents the voxel vertical coordinate of the i-th voxel feature of the laser scan point cloud.

[0149] Please refer to Figure 4 , the flowchart of the voxel feature clustering of the laser scan point cloud in the embodiment of the present invention. Specifically, by combining two clustering algorithms, DBSCAN and K-means, perform clustering analysis on the voxel features of the laser scan point cloud, and obtain a more accurate clustering result by iteratively optimizing the central cluster, so as to better understand the working state of the rotating equipment, which helps to improve the operation efficiency of the power system and reduce energy consumption. DBSCAN is a density clustering algorithm used to identify closely connected data points in space and divide them into clusters. DBSCAN can be used for preliminary clustering to obtain core points. A core point refers to a core area that contains at least a specified number of data points within a given radius. Since each core point may represent a cluster, the number of core points can be used to preliminarily determine the number of central clusters for clustering.

[0150] The core points obtained by preliminary clustering through DBSCAN are used as the initial central clusters of the K-means clustering algorithm, providing an initial cluster center position for the K-means algorithm; the Euclidean distance is used for clustering division. By calculating the Euclidean distance from the voxel features of the laser scan point cloud to each initial central cluster, each working operation information is assigned to the nearest central cluster. The average clustering error refers to the average of the Euclidean distances from all the data points included to this center. The average clustering error can be used to update the initial central clusters to improve the clustering effect. According to the calculated average clustering error, the position of the central clusters can be adjusted, thereby updating the central clusters. And the convergence condition is defined to determine whether the updated central clusters meet the conditions to determine whether to end the iteration. If the convergence condition is met, the clustering result is output.

[0151] Please refer to Figure 5 , the raster clustering result map of the embodiment of the present invention. First, traverse the raster map to find the first unmarked raster containing the voxel features of the laser scan point cloud, mark it as 1. Then, search for other rasters containing the voxel features of the laser scan point cloud in the neighborhood of this raster, and judge whether the central cluster of the voxel features of this laser scan point cloud is the same as that of the previous laser scan point cloud. If it is the same, also mark it as 1. If it is different, continue to search for other rasters, repeatedly mark the neighborhoods of the rasters until all the rasters in this connected domain are successfully marked. Record the positions of all the rasters in this connected domain. Then, repeat the above operations to find the remaining connected domains in the raster map and mark them as 2 and 3 according to the digital serial numbers. Finally, several connected regions are obtained, that is, the rasters with the same central clusters in the raster map. At this time, the voxel features of the laser scan point cloud stored in each raster in the same connected domain are merged to obtain the internal point cloud features;

[0152] Laser scan point cloud data usually contains a large amount of original point data, and these data may be affected by noise, external environment and equipment errors, thereby improving the overall quality of the point cloud data. This process can effectively reduce the errors caused by external factors, making the point cloud data more accurately reflect the actual situation of the tire surface. To ensure the accuracy of subsequent analysis, the preprocessing step first removes the noise points and abnormal points. The specific steps for preprocessing the laser scan point cloud are as follows:

[0153] S3.1.1: Calculate the y-axis coordinates of all the laser scan point clouds. If the y-axis coordinate of a certain point is less than 0, then delete this point from the laser scan point cloud;

[0154] S3.1.2: Traverse and query the laser scan point cloud to delete the laser scan point cloud data with the same radar rotation angle and echo distance;

[0155] S3.1.3: Set the fitting plane equation, calculate the coefficients of the fitting plane equation based on the laser scanned point cloud, and calculate the distances from all laser scanned point clouds to the fitting plane. Compare the distances from the laser scanned point clouds to the fitting plane with the plane distance threshold. If it is less than or equal to the plane distance threshold, save the laser scanned point cloud; if it is greater than the plane distance threshold, delete the laser scanned point cloud;

[0156] S3.1.4: Aggregate the laser scanned point cloud in the normal direction, adjust the positions and coordinates of the sampling points, so as to denoise the laser scanned point cloud.

[0157] Please refer to Figure 6 , the structural diagram of the feature fusion network in the embodiment of the present invention. The feature fusion network includes:

[0158] Input layer, establish a feature sequence according to the input parameters. The feature sequence includes an image point cloud feature sequence and a laser scanned point cloud feature sequence. The input layer receives external structure features and internal point cloud features as input parameters. These features respectively represent the surface structure and internal structure of the object to be modeled, and are key data extracted and processed from the previous steps. The external structure features reflect the external geometric shape and texture information of the object to be modeled, and the internal point cloud features reflect the internal structure and spatial layout of the object to be modeled;

[0159] Feature extraction layer, used to extract the feature vectors of the feature sequence. The feature extraction layer includes a geometric feature extraction module, a surface feature extraction module, an internal feature extraction and a spatial feature extraction module. The feature vectors include surface features, spatial features, internal features and geometric features;

[0160] Feature fusion layer, used to map the geometric features, the surface features, the internal features and the spatial features into a tuple sequence, and calculate the fusion features through pooling operations and residual connections.

[0161] Training the input parameters through the feature fusion network and calculating the fusion features specifically include:

[0162] Establish a grey differential equation through the input layer with the input parameters. The grey differential equation is a mathematical model used to describe the dynamic characteristics of a system, especially suitable for dealing with situations of uncertainty and insufficient information. Calculate the least squares parameters of the grey differential equation. Discretize the continuous solution obtained by solving the least squares method to calculate the fitting values of the feature sequence. These fitting values represent the image point cloud feature sequence and the laser scanned point cloud feature sequence generated based on the input parameters. The image point cloud feature sequence reflects the changes in the external features of the object to be modeled, and the laser scanned point cloud feature sequence reflects the changes in the internal features;

[0163] The feature extraction layer includes a geometric feature extraction module, a surface feature extraction module, an internal feature extraction module, and a spatial feature extraction module.

[0164] Please refer to Figure 7 , the structural diagram of the surface feature extraction module in the embodiment of the present invention. The surface feature extraction module is responsible for extracting and integrating surface shape features and surface structure features from the image point cloud feature sequence. It realizes the efficient extraction of complex three-dimensional surface features through hierarchical processing and high-level concatenation operations;

[0165] The surface feature extraction module includes a surface shape feature extraction sub-module, a surface structure feature extraction sub-module, and a feature concatenation sub-module;

[0166] The surface shape feature extraction sub-module includes a multi-layer perceptron network for extracting shape features from the image point cloud feature sequence. The multi-layer perceptron network (MLP) is a feed-forward neural network composed of multiple fully connected layers. Each layer of neurons receives the output of the previous layer of neurons and performs a non-linear transformation. In each layer of the MLP, first, an average pooling operation is performed on the image point cloud feature sequence. Average pooling is a down-sampling method that reduces the size of the feature map by calculating the average value of the input features and retains important statistical information. This operation helps to symmetrically process features and reduces the sensitivity to the asymmetry of the input data. After average pooling, a one-dimensional convolution operation is applied to the feature sequence. One-dimensional convolution is used to extract local patterns in the sequence and capture shape changes through a sliding window. The convolved feature sequence is normalized for standardization to improve the stability and convergence speed of training, describe the shape of the image point cloud feature sequence, and finally, activation is performed according to the ReLU activation function to obtain the surface shape features, which are used to describe the local and global information of the object surface in terms of surface shape.

[0167] The surface structure feature extraction sub-module includes a set of Gaussian kernels for learning. The Gaussian kernel function is a kernel method that can map input features to a higher-dimensional feature space through projection, thereby capturing more complex structural patterns. Here, the Gaussian kernel function is used to learn the radial distance, azimuth angle, and tilt angle of the image point cloud feature sequence. The Gaussian kernel function operates on the image point cloud feature sequence to calculate the radial distance (the distance from the reference point to the surface point), azimuth angle (the angle of the surface point relative to the reference axis), and tilt angle (the angle between the surface and the reference plane) of each surface. These parameters describe the spatial position and orientation of each surface and are important components of the surface structure features. After obtaining the radial distance, azimuth angle, and tilt angle, convolution operations are used to encode these features. Convolution operations can capture the local structural features of each surface, including local concavity and convexity, surface texture, etc., thereby encoding complex geometric structure information into feature vectors that can be used for subsequent processing to obtain surface structure features, which are used to describe the spatial relationships of various parts of the object surface.

[0168] The feature concatenation sub-module is used to concatenate the features obtained from the surface shape feature extraction sub-module and the surface structure feature extraction sub-module. The concatenation operation is to connect two feature vectors along the feature dimension to form a new, higher-dimensional feature vector. This operation retains all the information of the original features and combines them together to provide a more comprehensive feature representation. Finally, the output of the feature concatenation sub-module is the fused surface feature, which simultaneously includes the surface shape feature and the surface structure feature and can comprehensively describe the shape information and structure information of the object surface.

[0169] Please refer to Figure 8 , the structural diagram of the geometric feature extraction module in the embodiment of the present invention. The geometric feature extraction module is responsible for identifying and extracting basic geometric shape features from the image point cloud feature sequence. These geometric features include planes, curved surfaces, edges, etc., which constitute the basic geometric structure of the object surface. The function of the geometric feature extraction module lies not only in identifying these basic shapes but also in refining them into feature vectors that can be used for subsequent processing through precise mathematical and computational methods;

[0170] The geometric feature extraction module includes a geometric shape detection sub-module, a normal vector calculation sub-module, and a curvature estimation sub-module;

[0171] The geometric shape detection sub-module identifies the basic geometric shapes in the point cloud through RANSAC and least squares fitting, ensuring the reliability of geometric feature extraction. In point cloud data, planes are one of the most common geometric features, which can be achieved through the RANSAC algorithm. RANSAC is an iterative algorithm that continuously samples and verifies to find the plane equation that fits the largest part of the point cloud. The advantage of RANSAC is that it can accurately detect plane features even in the presence of noise and outliers. For non-planar parts, curves and surfaces are common geometric shapes. For surface detection, least squares fitting is a commonly used technique. This technique finds the best-fitting surface by minimizing the sum of the squared distances between the point cloud and the fitted surface, extracting the complex surface features in the point cloud.

[0172] The normal vector calculation sub-module is used to calculate the normal vector of each point. By smoothing, it ensures the consistency of the normal vector field, reduces the impact of noise on geometric analysis, and improves the stability of feature extraction. Normal vectors are important tools for describing geometric shapes. To calculate the normal vector, first, the neighborhood of each point needs to be determined. The KNN algorithm is used to find a preset number of nearest neighbor points for each point, and then these neighborhood points are used for plane fitting. Then, the covariance matrix of the neighborhood point set is calculated through the PCA algorithm and its eigenvalue decomposition is performed. The PCA method can extract the main directions, that is, the eigenvectors. The eigenvector corresponding to the smallest eigenvalue of the eigenvectors is the normal vector. The advantage of PCA lies in its sensitivity to the local point cloud structure, which can accurately capture the subtle changes on complex surfaces.

[0173] The curvature estimation sub-module is used to calculate the key parameters describing the degree of surface curvature. Through discrete methods and normal vector gradient analysis, the curvature estimation sub-module can accurately extract surface curvature information, helping to identify the bending features and local extreme points of the surface. The combined use of different curvature types enables the curvature estimation sub-module to conduct a more comprehensive analysis of the surface geometry, thereby improving the shape description ability of the model. By analyzing the change of the normal vector of each point according to the discrete method, the mean curvature of the object surface can be calculated. By calculating the rate of change of the direction of the normal vector field, the curvature of the surface in different directions can be determined. Through Gaussian function aggregation, the Gaussian curvature of the object surface is calculated. Gaussian curvature and mean curvature are the two most commonly used curvature types. Gaussian curvature reflects the overall bending of the surface, while mean curvature more reflects local features.

[0174] Please refer to Figure 9, Structural diagram of the spatial feature extraction module in an embodiment of the present invention. The spatial feature extraction module is used to extract the spatial layout features of an object from a laser scan point cloud feature sequence. These spatial features not only describe the relative positions and distances between points in the point cloud data, but also involve more advanced spatial relationships, such as proximity, symmetry, and global spatial structure. By extracting these spatial features, key spatial information can be provided for subsequent 3D modeling, analysis, and applications.

[0175] The spatial feature extraction module includes a spatial encoding sub-module, a spatial transformation sub-module, and a spatial feature aggregation sub-module.

[0176] The spatial encoding sub-module enhances the network's perception ability of the spatial relationship of the point cloud by introducing explicit spatial position information, making the feature extraction more accurate and comprehensive. First, the point cloud is normalized. The normalization process includes adjusting the coordinates of the point cloud data to a unified scale range (usually mapped to a unit cube) to ensure the comparability of point clouds from different scans. After normalization, the point cloud data is further standardized by moving the center of the point cloud to the coordinate origin and adjusting the distribution of the point cloud to have a unified scale. This step eliminates the inconsistencies caused by data offset and scale differences, and assigns position encoding to each point. By mapping the position information of the point into a high-dimensional space, the model can better perceive and process spatial information. The position encoding method usually uses sine and cosine functions to generate high-dimensional position vectors, which are associated with the spatial layout of the point cloud.

[0177] The spatial transformation sub-module performs spatial alignment on the point cloud data. It enables the point cloud to automatically adjust to a standard pose during processing through adaptive learning, thereby reducing the impact caused by changes in perspective and position. A small neural network is used to learn a 3x3 affine transformation matrix. This matrix can perform operations such as rotation, translation, and scaling on the input point cloud data, aligning the point cloud data in space. This process reduces spatial deformation while retaining the key spatial information in the point cloud data.

[0178] The spatial feature aggregation sub-module is used to extract local spatial features from point cloud data. By analyzing the relative positions and distances between points in the neighborhood, a high-dimensional spatial feature representation is constructed. The KNN algorithm is used to find the K nearest neighbors for each point, forming a local neighborhood. The KNN algorithm calculates the Euclidean distance between points in the high-dimensional space to determine the set of nearest neighbor points. The parameter K of KNN can be tuned through experiments to optimally balance local detail capture and computational complexity. After finding the neighborhood point set, by analyzing the relative positions and distances of these neighborhood points, a feature vector describing the local structure of the point cloud is extracted. Specific methods include calculating the centroid, maximum and minimum distances, local density, etc. of the neighborhood point set. These feature vectors encode the microscopic structure of the point cloud and are the specific manifestations of the local spatial layout. In the three-dimensional space of the point cloud, three-dimensional convolution operations are applied. By sliding the convolution kernel in space, local spatial patterns are captured. The 3D convolution network can identify complex features in the local space, such as volume structures, spatial symmetries, etc. Through multiple layers of 3D convolution networks, spatial features at different levels can be extracted. By hierarchically organizing features at different scales, a pyramid structure is formed. This structure can effectively fuse spatial features from different scales, enabling the network to simultaneously focus on local details and global layouts. At the top layer of the pyramid, feature fusion techniques are used to converge features at different scales into a unified spatial feature representation. Commonly used feature fusion methods include weighted average, max pooling, or using fully connected layers to integrate features at different scales.

[0179] After being processed by multi-view, multi-scale analysis and deep learning models, the internal feature extraction module generates a comprehensive feature vector, which represents the internal structural features of the object. This feature vector not only includes local and global geometric information but also reflects the material distribution and spatial relationships inside the object.

[0180] The feature fusion layer maps the spatial features, internal features, surface features, and geometric features to a tuple sequence [T(Q 1 ,K 1 ,J 1 ,P 1 ),T(Q 2 ,K 2 ,J 2 ,P 2 ),...,T(Q r ,K r ,J r ,P r )], where T(·) represents the tuple sequence, Q r represents the corresponding input of the r-th feature in the geometric features, K r represents the corresponding input of the r-th feature in the spatial features, J r represents the corresponding input of the r-th feature in the surface features, Pr Represents the corresponding input of the r-th feature in the internal features. Perform a dot product on the corresponding input features in each candidate sequence in the tuple order. Calculate the correlation score after the dot product of the candidate sequences through the Pearson correlation coefficient. Compress the correlation score according to the sigmoid function and compare it with the correlation threshold. If it is greater than the correlation threshold, retain the candidate sequence. If it is less than or equal to the correlation threshold, filter the candidate sequence. Collect the local features of the candidate sequences retained in the tuple order and enhance them. Concatenate the enhanced candidate sequences along the head sequence of the tuple order. Fuse the local features through residual connection. Perform a pooling operation on the feature vector after the residual connection fusion to obtain the fused feature. The calculation formula of the fused feature is as follows:

[0181] F fu = b{Con[En(T 1 ), En(T 2 ),..., En(T r )]},

[0182] where F fu represents the fused feature, Con[·] represents the residual connection function, En(·) represents the local enhancement function, and T r represents the r-th feature group in the tuple order;

[0183] The S5 imports the fused feature into the modeling software for processing, finally outputs the 3D model of the object to be modeled, and merges all the 3D models to construct a complete 3D substation model;

[0184] Before importing the fused feature into the modeling software, first, these features need to be converted into a data format recognizable by the modeling software. Import the converted fused feature data into the modeling software. During the import process, ensure that the scale, unit, and coordinate system of the model are consistent with the settings of the modeling software to avoid data distortion or misalignment. After completing the merger of each part of the model, integrate these models into a complete substation scene. Scene integration includes the layout, positioning, and scaling of the models, so that each component is in the correct position. The scene management tools of the modeling software can be used to organize and control the relationships and interactions between different models. During the scene construction process, assign physical properties such as materials, light reflectivity, and collision properties to different model components. These physical properties are crucial for simulation and interaction in the virtual space. By assigning the correct physical properties, the model not only has a visual sense of reality but also can correctly respond to environmental changes and user operations in the virtual space. After constructing the complete 3D substation model, perform integrity checking and verification. This includes checking whether there are defects, overlaps, or inconsistencies in the model and making adjustments and corrections through repair tools. Integrity checking ensures the quality and accuracy of the final model, enabling it to be smoothly applied in the virtual space.

[0185] Before filling the 3D model of the substation into the virtual space, it is necessary to set the environmental parameters of the virtual space. This includes setting the background, light source, terrain, etc. of the virtual space to make the 3D model match the virtual environment. By adjusting the environmental parameters, a more realistic virtual scene can be created, and the generated 3D model of the substation is imported into the virtual space. During the import process, adjust the position, angle, and scale of the model to ensure its correct placement in the virtual space. Virtual reality tools can be used to preview and adjust the position of the model to obtain the best display effect. Add interactive functions to the 3D model of the substation in the virtual space. This may include functions such as clicking to view details, dynamically simulating equipment operation, and remote operation. By integrating interactive functions, users can interact with the 3D model in the virtual space for simulation testing, operation training, or demonstration.

[0186] The digital service space visually displays the sensor data from the physical space and the 3D model from the virtual space through the Internet of Things platform. The digital service space includes equipment operation status monitoring, fault troubleshooting, fault warning, and maintenance guidance. The equipment operation status monitoring includes temperature monitoring, vibration monitoring, pressure monitoring, current monitoring, and flow monitoring. The equipment operation status monitoring also includes real-time analysis of all monitoring data, identifying abnormal data patterns, and calculating analysis results. The fault troubleshooting is used to predict faults through artificial intelligence algorithms based on historical data and analysis results. The fault warning is used to automatically send a warning signal when there are fault signs in the rotating equipment to remind the staff to carry out test, repair, and maintenance work. The maintenance guidance is used to generate maintenance operation guidance, including fault causes, recommended repair steps, and spare part requirements.

[0187] The temperature monitoring is used to monitor the temperature data of the rotating equipment according to the temperature sensors installed in the physical space, clean and preprocess the collected temperature data, draw a frequency distribution histogram of the temperature data, and output the temperature status analysis result of the rotating equipment according to the frequency distribution histogram.

[0188] The pressure monitoring is used to monitor the pressure data of the rotating equipment according to the pressure sensors installed in the physical space, preprocess the pressure data, including filtering and denoising, data smoothing, and outlier detection, to ensure the accuracy of subsequent analysis. Through time series analysis methods, detect whether there are abnormal pressure rising or falling trends, and decompose the pressure signal into different frequency components through wavelet analysis to detect high-frequency abnormal fluctuations.

[0189] The current monitoring is used to collect current signals in the power system of rotating equipment in real time according to current sensors installed in the physical space, such as Hall effect sensors or shunts. The current signals are rectified and smoothed to obtain stable DC signals. The frequency domain analysis of the current signals is carried out through short-time Fourier transform to identify different working states during system operation. The low-frequency components in the current may correspond to normal working states, while the high-frequency components may indicate abnormal load fluctuations in the system. By setting the detection threshold of the instantaneous current, the system can quickly identify current anomalies that may damage the equipment and take corresponding protection measures, and monitor and judge whether there are current anomalies in real time during equipment operation, so as to avoid potential risks such as equipment overload or short circuit.

[0190] The flow monitoring is used to collect flow data in the fluid system of rotating equipment according to flow sensors installed in the physical space, such as turbine flow meters, electromagnetic flow meters and ultrasonic flow meters. To ensure the accuracy of the data, the output of the flow sensor is first calibrated, including temperature compensation and zero drift correction. By calculating the input-output balance of the flow, it is possible to detect whether there are leakage or blockage phenomena in the system. For example, by comparing the readings of different flow meters, abnormal flow changes in a certain part of the pipeline or equipment can be identified, and potential problems can be inferred. The future flow trend is predicted through a time series prediction model to identify possible abnormal situations in advance. For example, if the predicted flow value exceeds or is lower than the preset safety range, the system can issue a warning in advance and initiate corresponding protection measures.

[0191] The vibration monitoring is used to collect vibration signals during the operation of rotating equipment according to vibration sensors installed in the physical space. The vibration signals are subjected to adaptive signal decomposition, and the kurtosis value of each component is calculated. The components are sorted in descending order according to the kurtosis value, and the top ten components are selected for signal reconstruction to eliminate vibration signal noise. The Fourier transform is performed on the vibration signals after signal reconstruction, and the frequency domain boundary of the Fourier spectrum of the vibration signals is divided through scale space transformation. The Fourier spectrum of the vibration signals is divided into frequency bands with different center frequencies, and each frequency band corresponds to an intrinsic mode function. Each intrinsic mode function is iterated to calculate the optimal solution under the condition that the sum of the bandwidths of each modal component is the smallest. The power spectrum is calculated according to the optimal solutions of each modal component to obtain the characteristic frequencies of the rolling bearing signals. The component with the largest energy in the power spectrum is selected as the inner ring signal of the rolling bearing, the component with the second largest energy as the outer ring signal of the rolling bearing, the component with the third largest energy as the rolling element signal of the rolling bearing, and the component with the second largest energy as the cage signal of the rolling bearing. The characteristic frequencies of the rolling bearing signals are compared with the theoretical frequencies of the rolling bearing, and the vibration signal analysis results are output. The vibration signal analysis results include the normal state of the bearing, the fault state of the bearing and the abnormal state of the bearing.

[0192] Example 2

[0193] Please refer to Figure 10 , the present invention provides an example: a maintenance system for a turning machine device based on digital twin technology, the system includes a device virtual module, a device operating state monitoring module, and a test and maintenance control module;

[0194] The device virtual module is used to perform three-dimensional visualization on the turning machine device in the substation, and the device virtual module is configured with a data processing strategy and a three-dimensional modeling strategy;

[0195] The device operating state monitoring module is used to monitor the turning machine device through a visualization interface, perform real-time analysis on the monitoring data, identify abnormal data patterns, and calculate the analysis results;

[0196] The test and maintenance control module is used to generate a test and maintenance plan according to the analysis results, convert the test and maintenance plan into a machine control instruction, and output the machine control instruction;

[0197] The test and maintenance control module is configured with a maintenance control strategy, and the maintenance control strategy analyzes the analysis results according to an intelligent decision-making algorithm, generates a test and maintenance plan, generates a maintenance instruction through a control algorithm, and controls a maintenance device to perform maintenance on the turning machine device through the maintenance instruction.

[0198] The device operating state monitoring module includes:

[0199] Device monitoring unit: used to monitor the turning machine device through sensors and obtain the monitoring data of the device;

[0200] Real-time data analysis unit: perform real-time analysis on all monitoring data, identify abnormal data patterns, and calculate the analysis results;

[0201] The device operating state monitoring module is configured with a data analysis strategy, and the data analysis strategy includes temperature analysis logic, vibration analysis logic, pressure analysis logic, current analysis logic, and flow analysis logic;

[0202] The temperature analysis logic is used to analyze the temperature data collected by the temperature sensor to determine whether there is an overheating fault;

[0203] The pressure analysis logic is used to analyze the pressure data collected by the pressure sensor, identify the trend and fluctuation of the pressure over time, and determine whether there is overpressure and pressure relief;

[0204] The flow analysis logic is used to analyze the flow data collected by the turbine flowmeter to determine whether there is a situation of flow exceeding the limit;

[0205] The temperature analysis logic, vibration analysis logic, pressure analysis logic, current analysis logic, and flow analysis logic are configured within the real-time data analysis unit.

[0206] The vibration analysis logic is used to analyze the vibration signals collected by the vibration sensor to determine whether there is abnormal vibration of the equipment bearings. Specifically, it includes:

[0207] Performing signal reconstruction on the vibration signals;

[0208] Performing Fourier transform on the vibration signals after signal reconstruction, and dividing the Fourier spectrum of the vibration signals into frequency bands with different center frequencies;

[0209] Iterating through each frequency band to calculate the optimal solution when the sum of the bandwidths satisfying each modal component is minimized;

[0210] Calculating the power spectrum based on the optimal solutions of each modal component, obtaining the characteristic frequencies of the rotating equipment bearing signals, comparing the characteristic frequencies of the rotating equipment bearing signals with the theoretical frequencies of the rotating equipment bearing signals, and outputting the analysis results of the vibration signals.

[0211] The current analysis logic is used to analyze the current signals collected by the current sensor to determine whether there is load startup and current overload. Specifically, it includes:

[0212] Rectifying and smoothing the current signals, performing frequency-domain analysis on the current signals through short-time Fourier transform, and extracting high-frequency components and low-frequency components;

[0213] Comparing the high-frequency components of the current signals with the current detection threshold. If it is greater than or equal to the current detection threshold, it indicates the presence of current overload;

[0214] Comparing the low-frequency components of the current signals with the initial current threshold. If it is greater than or equal to the initial current threshold, it indicates the presence of load startup.

[0215] The test and maintenance control module includes:

[0216] A scheme generation unit for processing the analysis results and generating a test and maintenance scheme with the minimum cost;

[0217] A scheme execution unit for converting the test and maintenance scheme into machine instructions executable by the system, and controlling the maintenance device to perform maintenance on the rotating equipment through the machine instructions;

[0218] An execution feedback unit for real-time monitoring of the operating state of the rotating equipment, comparing it with the expected maintenance results, and automatically adjusting the maintenance control strategy and machine control instructions according to the results of the real-time feedback;

[0219] The maintenance control strategy includes an optimal solution generation logic and a control logic. The optimal solution generation logic is configured in the solution generation unit, and the control logic is configured in the solution execution unit.

[0220] The optimal solution generation logic includes:

[0221] Construct a genetic algorithm, and define the objective function and constraints;

[0222] Use the analysis results and the equipment operation status as the input parameters of the genetic algorithm, and perform iterative optimization on the input parameters through the genetic algorithm to output the optimal solution of the objective function;

[0223] Use the optimal solution of the objective function as the test maintenance plan.

[0224] The equipment virtual module includes:

[0225] A three-dimensional virtual model construction unit, which is used to construct a three-dimensional virtual model of the substation and perform data mapping and interaction between the physical space, virtual space, and digital service space;

[0226] A sensor data acquisition unit, which is used to obtain the external image of the object to be modeled and the laser scanning point cloud of the object to be modeled through a variety of sensor components;

[0227] A data processing unit, which is used to process the external data and internal data of the object to be modeled and perform three-dimensional visualization of the rotating equipment in the substation.

[0228] The three-dimensional virtual model construction unit includes:

[0229] A physical space subunit: including physical devices and their layouts, and collecting physical data of the devices through sensors;

[0230] A virtual space subunit: digitally representing the devices and environment in the physical space, including the three-dimensional models of the devices, virtual environments, and simulation data;

[0231] A digital service space subunit: used to provide digital services such as real-time monitoring, data analysis, and maintenance guidance;

[0232] The data processing strategy includes an image processing logic and a point cloud processing logic. The image processing logic is used to extract the surface features of the rotating equipment through image processing techniques, and the point cloud processing logic is used to extract the internal features of the rotating equipment;

[0233] The data processing unit includes:

[0234] An image processing subunit: processing the external data and extracting the external structural features;

[0235] Point cloud processing subunit: Process internal data and extract internal point cloud features;

[0236] Feature fusion subunit: Fuse external structural features and internal point cloud features and generate a three-dimensional model of the object to be modeled;

[0237] Model merging subunit: Merge multiple three-dimensional models to construct a complete three-dimensional model of the substation;

[0238] The three-dimensional modeling strategy includes a feature fusion logic for fusing external structural features and internal point cloud features.

[0239] The image processing logic is configured in the image processing subunit, and the point cloud processing logic is configured in the point cloud processing subunit;

[0240] The feature fusion logic is configured in the feature fusion subunit.

[0241] Although the embodiments of the present invention have been shown and described above, it can be understood that the above embodiments are exemplary and should not be construed as limiting the present invention. Those of ordinary skill in the art can make changes, modifications, substitutions, and variations to the above embodiments within the scope of the present invention.

Claims

1. A transfer equipment maintenance system based on digital twin technology, characterized in that: The system includes an equipment virtual module, an equipment operation status monitoring module and a test and maintenance control module; The equipment virtualization module is used to perform three-dimensional visualization of the transfer equipment in the substation, and the equipment virtualization module is configured with a data processing strategy and a three-dimensional modeling strategy, wherein the data processing strategy includes image processing logic and point cloud processing logic; The equipment operation status monitoring module monitors the transfer equipment through a visual interface, and performs real-time analysis on the monitoring data, identifies abnormal data patterns, and calculates the analysis results; The test and maintenance control module generates a test and maintenance plan according to the analysis result, converts the test and maintenance plan into a machine control instruction, and outputs the machine control instruction; The test and maintenance control module is provided with a maintenance control strategy, which analyzes the analysis results according to an intelligent decision-making algorithm, generates a test and maintenance plan, and generates maintenance instructions through a control algorithm, and controls the maintenance device to perform maintenance on the transfer equipment through the maintenance instructions; The image processing logic comprises: Compressing the external image of the object to be modeled to obtain a compressed reconstructed image; Perform camera calibration on the image sensor, obtain the extrinsic parameter matrix and intrinsic parameter matrix of the image sensor, perform coordinate transformation on the compressed reconstructed image according to the extrinsic parameter matrix and the intrinsic parameter matrix, and transform the pixels of the compressed reconstructed image from the camera coordinate system to the world coordinate system; Calculate the global mean of all pixels of the compressed reconstructed image after coordinate transformation, perform weighted sampling through the upsampling function, and calculate the global feature map; Compare the confidence of the pixel points of the global feature map with the confidence threshold, retain the pixel points greater than the confidence threshold, and obtain the surface feature map; Performing regional cutting on the surface feature map, extracting the regional temporal features after the surface feature map is cut according to the temporal feature extraction function, and extracting the regional spatial features after the surface feature map is cut through the attention mechanism; The regional temporal features and regional spatial features of the surface feature map are fused to obtain the bilinear regional features. The bilinear regional features of all regions are integrated through the pooling function to calculate the external structural features.

2. According to claim 1, a transfer equipment maintenance system based on digital twin technology is characterized in that: The equipment operation status monitoring module comprises: Equipment monitoring unit: monitors the transfer equipment through sensors and obtains equipment monitoring data; Real-time data analysis unit: performs real-time analysis on all monitoring data, identifies abnormal data patterns and calculates analysis results; The equipment operation status monitoring module is configured with a data analysis strategy, which includes temperature analysis logic, vibration analysis logic, pressure analysis logic, current analysis logic and flow analysis logic; The temperature analysis logic is used to analyze the temperature data collected by the temperature sensor to determine whether there is an overheating fault; The pressure analysis logic is used to analyze the pressure data collected by the pressure sensor, identify the trend and fluctuation of pressure over time, and determine whether there is overpressure and pressure relief; The flow analysis logic is used to analyze the flow data collected by the turbine flow meter to determine whether there is a flow exceeding the limit; The temperature analysis logic, vibration analysis logic, pressure analysis logic, current analysis logic and flow analysis logic are configured in the real-time data analysis unit.

3. According to claim 2, a transfer equipment maintenance system based on digital twin technology is characterized in that: The vibration analysis logic is used to analyze the vibration signal collected by the vibration sensor to determine whether there is abnormal vibration of the equipment bearing, specifically including: Reconstruct the vibration signal; Performing Fourier transform on the reconstructed vibration signal, dividing the Fourier spectrum of the vibration signal into frequency bands with different center frequencies; Iterate each frequency band and calculate the optimal solution that satisfies the minimum sum of bandwidths of each modal component. The power spectrum is calculated according to the optimal solution of each modal component, the characteristic frequency of the bearing signal of the transfer equipment is obtained, the characteristic frequency of the bearing signal of the transfer equipment is compared with the theoretical frequency of the bearing signal of the transfer equipment, and the vibration signal analysis result is output.

4. According to claim 2, a transfer equipment maintenance system based on digital twin technology is characterized in that: The current analysis logic analyzes the current signal collected by the current sensor to determine whether there is load startup and current overload, specifically including: The current signal is rectified and smoothed, and the frequency domain analysis of the current signal is performed through short-time Fourier transform to extract high-frequency and low-frequency components; Compare the high frequency component of the current signal with the current detection threshold. If it is greater than or equal to the current detection threshold, it indicates that there is a current overload. The low-frequency component of the current signal is compared with the current initial threshold value. If it is greater than or equal to the current initial threshold value, it indicates that a load is started.

5. According to claim 1, a transfer equipment maintenance system based on digital twin technology is characterized in that: The test and maintenance control module comprises: A solution generation unit is used to process the analysis results and generate a minimum cost test and maintenance solution; A scheme execution unit, used to convert the test and maintenance scheme into a machine instruction executable by the system, and control the maintenance device to perform maintenance on the transfer equipment through the machine instruction; The execution feedback unit is used to monitor the operating status of the transfer equipment in real time and compare it with the expected maintenance results. According to the real-time feedback results, the maintenance control strategy and machine control instructions are automatically adjusted; The maintenance control strategy includes an optimal solution generation logic and a control logic. The optimal solution generation logic is configured in the solution generation unit, and the control logic is configured in the solution execution unit.

6. According to claim 5, a transfer equipment maintenance system based on digital twin technology is characterized in that: The optimal solution generation logic includes: Construct a genetic algorithm and define the objective function and constraints; The analysis results and the equipment operation status are used as the genetic algorithm input parameters, the input parameters are iteratively optimized by the genetic algorithm, and the optimal solution of the objective function is output; The optimal solution of the objective function is used as the test and maintenance plan.

7. According to claim 1, a transfer equipment maintenance system based on digital twin technology is characterized in that: The device virtual module includes: A three-dimensional virtual model building unit, used to build a three-dimensional virtual model of the substation and perform data mapping and interaction between the physical space, virtual space and digital service space; A sensor data acquisition unit, which acquires an external image of the object to be modeled and a laser scanning point cloud of the object to be modeled through a variety of sensor components; The data processing unit is used to process the external data and internal data of the object to be modeled and to perform three-dimensional visualization of the transfer equipment in the substation.

8. According to claim 7, a transfer equipment maintenance system based on digital twin technology is characterized in that: The three-dimensional virtual model building unit comprises: Physical space unit: includes physical equipment and its layout, and collects physical data of equipment through sensors; Virtual space unit: digital representation of the equipment and environment of the physical space, including the 3D model of the equipment, the virtual environment and simulation data; Digital service space unit: used to provide digital services such as real-time monitoring, data analysis and maintenance guidance.

9. According to claim 8, a transfer equipment maintenance system based on digital twin technology is characterized in that: The data processing unit comprises: Image processing subunit: processes external data and extracts external structural features; Point cloud processing subunit: processes internal data and extracts internal point cloud features; Feature fusion subunit: fuses external structural features and internal point cloud features to generate a 3D model of the object to be modeled; Model merging subunit: merge multiple 3D models to build a complete 3D model of the substation; The data processing strategy includes image processing logic and point cloud processing logic. The image processing logic extracts surface features of the transfer equipment through image processing technology, and the point cloud processing logic is used to obtain internal features of the transfer equipment. The three-dimensional modeling strategy includes feature fusion logic, and the feature fusion logic is used to fuse external structural features and internal point cloud features; The image processing logic is configured in the image processing subunit, and the point cloud processing logic is configured in the point cloud processing subunit; The feature fusion logic is configured in the feature fusion subunit.

10. The transfer equipment maintenance system based on digital twin technology according to claim 9 is characterized in that: The calculation formula of the external structural characteristics is: Among them, Gb fd represents the external structural features, a represents the cut area of ​​a single surface feature map, A represents the total number of cut areas of the surface feature map, b{·} represents the pooling function, T f (·) represents the time feature extraction function, hb fi represents the surface feature map, * represents the fusion operation, Ex[·] represents the excitation function in the attention mechanism, W represents the horizontal length of the surface feature map, H represents the vertical length of the surface feature map, n represents the unit horizontal length of the surface feature map, m represents the unit vertical length of the surface feature map, v a Represents the convolution kernel corresponding to the cut area.

11. The transfer equipment maintenance system based on digital twin technology according to claim 9 is characterized in that: The point cloud processing logic includes: Preprocessing the laser scanning point cloud according to the laser scanning point cloud of the internal parts of the object to be modeled; Find the most suitable 3D plane through plane fitting, perform coordinate transformation on the pre-processed 3D point cloud data, and transfer the laser scanning point cloud from the radar coordinate system to the world coordinate system; The world coordinate system is divided into grids at equal intervals, the laser scanning point cloud is represented by voxels, and the voxel index of the laser scanning point cloud is projected to the corresponding grid, the points in the voxel are connected, the local feature information is enhanced, and the voxel features of the laser scanning point cloud are obtained; The laser scanning point cloud voxel features are clustered according to the clustering algorithm, and the grids where the laser scanning point cloud voxel features are located under the same central cluster are merged to extract the internal point cloud features. The calculation formula of the internal point cloud features is: Among them, T a represents the internal point cloud features corresponding to the voxel features of the laser scanning point cloud in the same central cluster after clustering, i represents a single voxel feature of the laser scanning point cloud in the central cluster, M represents the total number of voxel features of the laser scanning point cloud in the central cluster, and η i represents the eigenvalue of the voxel feature of the i-th laser scanning point cloud, d represents the clustering threshold corresponding to the central cluster, r2 represents the farthest distance scanned by the laser scanner, r1 represents the closest distance scanned by the laser scanner, and p i_x represents the voxel horizontal coordinate of the i-th laser scanning point cloud voxel feature, p i_y The voxel ordinate represents the feature of the i-th laser scanning point cloud voxel, p i_z Represents the voxel vertical coordinate of the i-th laser scanning point cloud voxel feature.

12. The transfer equipment maintenance system based on digital twin technology according to claim 11, characterized in that: The cluster analysis comprises: Preliminarily clustering the voxel features of the laser scanning point cloud by using the DBSCAN clustering algorithm to obtain core points; The core points after DBSCAN clustering are used as the initial center clusters of the K-means clustering algorithm, and the Euclidean distances from the voxel features of the laser scanning point cloud to the corresponding initial center clusters are calculated; Clustering is performed according to the Euclidean distance, and the average clustering error of the initial central cluster is calculated. The central cluster is updated according to the average clustering error, and it is determined whether the central cluster meets the convergence condition after the update. If the convergence condition is not met, the central cluster is continued to be updated. If the convergence condition is met, the clustering result is output.

13. The transfer equipment maintenance system based on digital twin technology according to claim 9 is characterized in that: The feature fusion logic includes: Converting the external structural features and the internal point cloud features into a feature sequence, wherein the feature sequence includes an image point cloud feature sequence and a laser scanning point cloud feature sequence; Constructing geometric feature extraction module, surface feature extraction module, internal feature extraction module and spatial feature extraction module; Processing the image point cloud feature sequence by the geometric feature extraction module and the surface feature extraction module to extract geometric features and surface features; Processing the laser scanning point cloud feature sequence by the internal feature extraction module and the spatial feature extraction module to extract internal features and spatial features; The geometric features, the surface features, the internal features and the spatial features are mapped into a tuple order, and fusion features are calculated through pooling operations and residual connections.

14. The transfer equipment maintenance system based on digital twin technology according to claim 13, characterized in that: The surface feature extraction module is used to extract and integrate surface shape features and surface structure features from the image point cloud feature sequence, and the surface feature extraction module includes a surface shape feature extraction submodule, a surface structure feature extraction submodule and a feature cascade submodule; The geometric feature extraction module is used to identify and extract geometric shape features from the image point cloud feature sequence, and the geometric feature extraction module includes a geometric shape detection submodule, a normal vector calculation submodule and a curvature estimation submodule; The spatial feature extraction module is used to extract the spatial layout features of the object from the laser scanning point cloud feature sequence, and the spatial feature extraction module includes a spatial encoding submodule, a spatial transformation submodule and a spatial feature aggregation submodule; The internal feature extraction module is used to extract the internal material distribution features of the object from the laser scanning point cloud feature sequence.

15. The transfer equipment maintenance system based on digital twin technology according to claim 14, characterized in that: The calculation formula of the fusion feature is: F fu =b{With[In(T1),In(T2),...,In(T r )]}, Among them, F fu represents fusion features, Con[·] represents residual connection function, En(·) represents local enhancement function, T r represents the rth feature group in the tuple sequence, and b{·} represents the pooling function.

16. A method for maintenance of transfer equipment based on digital twin technology, the method being implemented based on a transfer equipment maintenance system based on digital twin technology as claimed in any one of claims 1 to 15, characterized in that: The method comprises: Constructing a three-dimensional virtual model of a substation, wherein the three-dimensional virtual model of the substation includes a physical space, a virtual space, and a digital service space; Determine an object to be modeled according to the physical space, and obtain sensor data of the object to be modeled through a sensor component; Processing the sensor data, constructing a three-dimensional model of a substation, and filling the three-dimensional model of the substation into the virtual space; Mapping the physical space and the virtual space to the digital service space, and displaying the operation status, fault location information and test and maintenance plan of the equipment in real time through the digital service space; A maintenance instruction is generated according to the test maintenance plan, and the maintenance device is controlled by the maintenance instruction to perform maintenance on the transfer equipment.

Images