A substation operation scheme compiling method based on a three-dimensional model

Abstract

The application discloses a kind of substation operation scheme based on three-dimensional model, it is related to electric power operation maintenance technical field, including: acquisition multi-source data, construct three-dimensional digital twin and access real-time data, form digital twin platform;Query maintenance task and carry out text structured processing, form maintenance equipment set;Carry out special analysis and provide special analysis result;Maintenance scene is shown in digital twin platform, generates and exports maintenance operation scheme;After maintenance operation, back transmission on-site maintenance data, update digital twin platform data.The application is mapped with substation synchronously by constructing three-dimensional digital twin, and is dynamically updated, so that maintenance operation scheme is prepared based on real-time on-site state, solve the problem that traditional three-dimensional visualization system is static display and is disjointed with field, convert traditional dependence artificial experience, complicated and easy-to-mistake step into automatic algorithm calculation, greatly reduce the time required for special analysis generation, reduce operation and maintenance cost.

Description

Technical Field

[0001] This invention relates to the field of power operation and maintenance technology, specifically a method for compiling substation operation plans based on a three-dimensional model. Background Technology

[0002] Existing 3D visualization systems often remain at the stage of static 3D display or are limited to visualization of a single data source. They can only present pre-built models of substations or transmission lines and cannot drive the models to change in real time according to on-site conditions. 3D visualization systems only connect to one of the following: PMS ledgers, SCADA telemetry, or video streams. This leads to a disconnect between the screen and the actual situation. When operators need to compare information from multiple sources such as load, defects, and weather, they need to switch back and forth between multiple screens, which is time-consuming and labor-intensive. Moreover, the system does not connect to real-time operational data, so the model cannot reflect dynamic changes such as real-time opening and closing of switches, disconnector temperatures, and conductor sag. At the same time, existing 3D visualization systems generally lack outage topology verification and automatic safety distance verification functions. When the dispatcher selects equipment for maintenance on the screen, the system cannot automatically deduce the impact range, verify whether dual-power users have been missed, or dynamically calculate risk values ​​such as clearance, crossing, and crane boom extension based on the real-time coordinates of energized parts and the 3D positioning of operators.

[0003] In summary, existing 3D visualization systems lack an integrated operational plan development system on a high-precision digital twin platform: In the power outage area topology verification stage, manual review of single-line diagrams and verification of user lists are still required; in the real-time safety distance verification stage, operators rely on experience for visual estimation or use rangefinders to measure point by point, resulting in large errors and low efficiency; in the one-click vehicle path planning stage, cranes and boom lifts often rely on driver memory to enter the site, frequently being forced to turn back due to height restrictions or insufficient turning radius; and in the post-operation data write-back unit, inspection results, defect photos, and measured distances cannot be automatically archived to the twin model, requiring repeated surveys in the next round of operations.

[0004] Based on this, a method for compiling substation operation plans based on three-dimensional models is now provided, which can eliminate the drawbacks of existing technical solutions. Summary of the Invention

[0005] The purpose of this invention is to provide a method for compiling substation operation plans based on a three-dimensional model, so as to solve the problem that existing three-dimensional visualization systems in the background art have not formed an integrated operation plan compilation system.

[0006] To achieve the above objectives, the present invention provides the following technical solution:

[0007] A method for developing substation operation plans based on a 3D model, specifically including the following steps:

[0008] Step S1: Collect and fuse multi-source data from the substation, construct a three-dimensional digital twin, and connect it to real-time data to form a digital twin platform that supports multi-terminal interaction;

[0009] Step S2: Query the maintenance task and parse the key information in the maintenance task, extract the key information and perform text structuring processing, match the corresponding device node in the digital twin platform through the device ID to form a maintenance equipment set, and output the maintenance constraints.

[0010] Step S3: Conduct a special analysis on the set of maintenance equipment in the digital twin platform and provide the special analysis results. The special analysis includes remote surveying and mapping, power outage area topology analysis, safety measure simulation and vehicle route planning.

[0011] Step S4: Integrate the special analysis results, overlay environmental factors, display the maintenance scenario in the digital twin platform, and generate and output the maintenance operation plan;

[0012] Step S5: After the maintenance work is completed, transmit the on-site maintenance data back and update the data on the digital twin platform;

[0013] The digital twin platform comprises a perception layer, a twin layer, an algorithm layer, and an interaction layer. The perception layer consists of a lidar module and an oblique photography module, used to collect raw data from the substation. The twin layer includes an equipment node encoding module, a time series caching module, and a real-time data mapping module, used to assign a unique identifier to each device through equipment node encoding and to associate real-time equipment operation data and environmental data with the three-dimensional digital twin through real-time data mapping. The algorithm layer includes a remote survey module, a power outage analysis module, a safety measure pre-simulation module, a vehicle routing module, and a deviation calculation module, used to conduct specialized analysis on the set of maintenance equipment. The interaction layer includes a three-dimensional roaming module, a multi-layer overlay module, a solution packaging module, and a version evolution module, used to receive and output data and instructions and provide operational functions to users. The digital twin platform is equipped with a connection port for data access operations from meteorological APIs, power grid resource business platforms, and defect databases.

[0014] Preferably, step S1 specifically includes:

[0015] Step S1-1: The entire area of ​​the substation is scanned by combining laser point cloud with oblique photography to generate high-density point cloud data. Based on the high-density point cloud data, GIS coordinate registration is performed to unify the point cloud data, image data, and wiring diagram into the same coordinate system.

[0016] Step S1-2: When constructing a 3D digital twin, the 3D entity is meshed, LOD lightweighted, and WebGL loadable processed, and device node encoding is performed to generate a unique ID for each device as a data anchor point.

[0017] Steps S1-3: Access the real-time measurement center data through the gateway. The real-time measurement center data includes real-time power, temperature, defects, hidden dangers, and historical maintenance work orders. Construct an equipment defect database based on the defects and historical maintenance work orders.

[0018] Steps S1-4: The digital twin platform interacts with web and mobile terminals and supports viewing in a visual format on web and mobile terminals.

[0019] Preferably, step S2 specifically includes:

[0020] Step S2-1: Query and obtain the task ID, power outage period, equipment list, and original text of the three measures and one plan for the maintenance task. The three measures and one plan include organizational measures, technical measures, safety measures, and work plan.

[0021] Step S2-2: Segment the task text of the maintenance task, obtain key information, and generate structured text according to the text format template;

[0022] Step S2-3: Query the device nodes that correspond one-to-one with the device ID in the digital twin platform to form a set of maintenance devices with coordinates;

[0023] Step S2-4: Output maintenance constraints, which include a set of maintenance equipment, a timeline, a procedure template, and a list of tools and equipment.

[0024] Preferably, step S3 specifically includes:

[0025] Step S3-1: Perform remote surveying and mapping operations, and calculate the horizontal distance, vertical distance and spatial distance in real time in the three-dimensional digital twin;

[0026] Step S3-2: Perform a topology analysis of the power outage area. The input is the adjacency matrix of the primary wiring diagram. The connected subgraph algorithm is used to calculate the power outage area and the live area, and the compliance of the power outage area is indicated.

[0027] Step S3-3: Perform a safety measure pre-simulation operation to generate a three-dimensional virtual fence from the electrified boundary and generate a safety measure layout diagram that includes fence coordinates, length, sign position and orientation.

[0028] Step S3-4: Perform vehicle path planning, construct a passable network, plan and generate the shortest collision-free path, and verify the turning radius and safety distance. The output is a vehicle attitude sequence with timestamps. The passable network includes the road centerline, height restriction data, and obstacle data.

[0029] Preferably, step S3-1 specifically includes:

[0030] Input two 3D measurement points and , The starting point for measurement is at coordinates [0, 1]. , The endpoint of the measurement is located at coordinates [omitted]. Detection and If a measurement point is valid, the horizontal distance, vertical distance, and spatial distance are calculated. When the survey line connecting two valid three-dimensional measurement points crosses the area of ​​live equipment in three-dimensional space, the shortest safe distance check is automatically triggered.

[0031] The formula for calculating horizontal distance is: ;

[0032] The formula for calculating vertical distance is: ;

[0033] The formula for calculating spatial distance is: ;

[0034] in, Horizontal distance Vertical distance For spatial distance, , , These are two three-dimensional measurement points at... , , Coordinate difference on the axis , , .

[0035] Preferably, step S3-2 specifically includes:

[0036] The adjacency matrix of the first-order wiring diagram is used as input, and the adjacency matrix is... phalanx, Given the total number of nodes in a wiring diagram, if the matrix elements... , representing a node With nodes There is a direct electrical connection between the matrix elements. , representing a node With nodes There is no direct connection between them;

[0037] The connected subgraph algorithm is used in conjunction with the on / off state of the equipment switch to calculate the power outage area and the energized area. The power outage area is the set of nodes that have lost connection with all power sources, and the energized area is the set of nodes that are still connected to the power sources.

[0038] Compliance verification of power outage areas is performed using three indicators: internal connectivity, ring network isolation, and scale limitation. Internal connectivity is achieved by generating a power outage area induced subgraph and detecting its connectivity. Ring network isolation is achieved by determining whether the power outage area and the energized area form a loop. Scale limitation is achieved by comparing the number of nodes in the power outage area with a preset threshold. If any one of the verifications fails, an alarm is triggered.

[0039] Preferably, step S3-3 specifically includes:

[0040] According to the enclosure of the electrical equipment, respectively , , The minimum and maximum values ​​on the axis, along with the preset safe expansion distance, are used to calculate the two-dimensional expansion corner point, and then the two-dimensional expansion corner point is moved along... Shaft lifting preset height This yields a sequence of three-dimensional virtual fence nodes, generating a closed three-dimensional virtual fence.

[0041] The total length of the three-dimensional virtual fence is calculated by accumulating the Euclidean distance between adjacent nodes using a linear approximation method. The fence is divided into equal parts, and signs are placed at the midpoint of each fence segment. The position and orientation of the signs are determined, and a safety layout diagram containing fence coordinates, length, sign position and orientation is generated.

[0042] Traverse any node in the 3D virtual fence, check whether the distance from the node to the energized equipment meets the safety threshold. If it does not meet the threshold, store the node in the alarm set and trigger an alarm.

[0043] Preferably, steps S3-4 specifically include:

[0044] Filter the set of feasible nodes whose node height is greater than or equal to the minimum height of the vehicle and whose distance from the node to the obstacle is greater than or equal to the safe distance threshold of the vehicle. Filter the set of feasible edges whose edge length is greater than or equal to the minimum turning radius of the vehicle and whose edge lines do not intersect with the obstacles. Construct a passable network that includes the road centerline, height restriction data and obstacle data.

[0045] Find the shortest collision-free path from the starting point to the ending point in a traversable network, wherein the shortest collision-free path satisfies the minimum turning radius of the vehicle and the safety distance threshold.

[0046] Based on the first node time, timestamps are assigned to path nodes according to vehicle speed to generate a vehicle attitude sequence of running trajectory with timestamps. The vehicle attitude sequence includes node spatial coordinates, vehicle quaternion orientation and corresponding time.

[0047] Preferably, step S4 specifically includes:

[0048] Step S4-1: Perform real-time layer overlay operation in any three-dimensional scene displayed on the digital twin platform. The layers include survey map, power outage area map, safety measure layout map, and vehicle route map.

[0049] Step S4-2: Perform an environment overlay operation on the same 3D scene after real-time layer overlay. The environment includes real-time weather, maintenance distribution map, and risk level.

[0050] Step S4-3: Generate a maintenance operation plan through the digital twin platform. The maintenance operation plan includes a structured "three measures and one plan", work order, operation order, 3D scene screenshot, vector DWG image and executable browsing file.

[0051] Preferably, step S5 specifically includes:

[0052] Step S5-1: Transmit on-site maintenance data through inspection equipment, including mobile terminals, drones and inspection robots, and the on-site maintenance data includes GPS trajectory, photos and defect elimination results.

[0053] Step S5-2: Map the GPS track and photos one by one to the digital twin platform. The digital twin platform automatically refreshes the device defect database and ledger status.

[0054] Step S5-3: Compare the actual maintenance data with the maintenance operation plan, calculate the deviation index between the two, and adjust the path planning weight and safety threshold of the subsequent maintenance operation plan.

[0055] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0056] 1. This invention integrates multi-source data such as lidar and oblique photography through the perception layer, and accesses real-time measurement center data through the twin layer to construct a three-dimensional digital twin that is synchronously mapped and dynamically updated with the substation. This enables maintenance operation plans to be compiled based on real-time on-site conditions, solving the problem of static display and disconnection from the site in traditional three-dimensional visualization systems, and improving the accuracy of the plans.

[0057] 2. This invention transforms the traditional, tedious, and error-prone steps that rely on manual experience into automated algorithmic calculations. This significantly reduces the time required for generating specific analyses such as power outage range, safety barriers, and vehicle paths. The method includes a power outage range topology algorithm and a safety distance verification operation, which can reduce errors in power outage boundaries, energized parts, and barrier placement. It also performs real-time collision detection on vehicle paths to prevent accidental entry into energized areas or exceeding height or width limits, thereby reducing human-caused accidents such as misoperation and accidental entry into gaps. This reduces the number of manual surveys, repeated site visits, and power outage tests, shortens the average power outage time per maintenance, and lowers operation and maintenance costs.

[0058] 3. This invention enables the digital twin platform to automatically update the equipment status in the digital twin by transmitting and comparing on-site data after the operation, and to calculate the deviation index between the actual situation and the plan. This is used to optimize the generation parameters of subsequent operation plans, and the plan package can be automatically packaged and exported for easy viewing in multiple ways, thus improving the flexibility of use. Attached Figure Description

[0059] Figure 1 This is a schematic diagram of the steps of the method of the present invention.

[0060] Figure 2 This is a flowchart illustrating step S1 of the present invention.

[0061] Figure 3 This is a flowchart illustrating step S2 of the present invention.

[0062] Figure 4 This is a flowchart illustrating step S3 of the present invention.

[0063] Figure 5 This is a flowchart illustrating step S4 of the present invention.

[0064] Figure 6 This is a flowchart illustrating step S5 of the present invention.

[0065] Figure 7 This is a schematic diagram of the architecture of the digital twin platform of the present invention.

[0066] Figure 8 This is a schematic diagram illustrating the process of constructing a digital twin platform and accessing data in real time according to the present invention.

[0067] Figure 9 This is a schematic diagram of the remote surveying and mapping operation process of the present invention.

[0068] Figure 10 This is a flowchart illustrating the power outage range topology analysis operation of the present invention.

[0069] Figure 11 This is a schematic diagram of the safety measure pre-operation process of the present invention.

[0070] Figure 12 This is a flowchart illustrating the vehicle routing operation of the present invention. Detailed Implementation

[0071] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments.

[0072] Among them, such as Figures 1-12 As shown, a method for compiling substation operation plans based on a three-dimensional model includes the following steps:

[0073] Step S1: Collect and fuse multi-source data from the substation, construct a three-dimensional digital twin, and connect it to real-time data to form a digital twin platform that supports multi-terminal interaction;

[0074] Step S2: Query the maintenance task and parse the key information in the maintenance task, extract the key information and perform text structuring processing, match the corresponding equipment node in the digital twin platform through the equipment ID to form a set of maintenance equipment, and output the maintenance constraints.

[0075] Step S3: Conduct a special analysis on the maintenance equipment set in the digital twin platform and provide the special analysis results. The special analysis includes remote surveying and mapping, power outage area topology analysis, safety measure simulation and vehicle route planning.

[0076] Step S4: Integrate the results of the special analysis, overlay environmental factors, display the maintenance scenario in the digital twin platform, and generate and output the maintenance operation plan;

[0077] Step S5: After the maintenance work is completed, transmit the on-site maintenance data back and update the data on the digital twin platform;

[0078] like Figure 7As shown, the digital twin platform is a comprehensive system integrating data acquisition, modeling, analysis, interaction, and feedback. It is used to construct, operate, and maintain a 3D digital twin of a substation. The 3D digital twin is the data model within the digital twin platform. Through equipment node coding and real-time data mapping, equipment status, environmental data, etc., are associated with the 3D model, achieving synchronous mapping with the physical substation. The digital twin platform includes a perception layer, a twin layer, an algorithm layer, and an interaction layer. The perception layer consists of a LiDAR module and an oblique photography module, used to collect raw substation data, including equipment 3D structural data (LiDAR, oblique photography) and historical equipment problem data (defect database). The twin layer includes an equipment node coding module, a time-series caching module, and a real-time data mapping module. It assigns a unique identifier to each device through equipment node coding and, combined with real-time data mapping, maps real-time equipment operating data and environmental data. The data is linked to a 3D digital twin, facilitating the operations in steps S1 and S2. The algorithm layer includes a remote survey module, a power outage analysis module, a safety measure pre-simulation module, a vehicle routing module, and a deviation calculation module, used to conduct specialized analysis on the set of maintenance equipment, facilitating the specialized analysis operation in step S3. The interaction layer includes a 3D roaming module, a multi-layer overlay module, a solution packaging module, and a version evolution module, used to receive and output data and instructions and provide operational functions to users. The digital twin platform has a connection port for data access operations from meteorological APIs, power grid resource business platform, and defect database. It can uniformly access the raw data collected by the perception layer and the maintenance tasks, equipment ledgers, and other data from the power grid resource business platform to the twin layer to achieve data aggregation. The equipment defect database is a dedicated equipment defect database of the digital twin platform built based on defects in the defect database and historical maintenance work orders.

[0079] Specifically, such as Figure 8As shown, the specific process of building a digital twin platform and accessing real-time data includes: After the digital twin operation carrier starts (BeginPlay), it initializes the substation ID (InitStationId), sends an API request to the power grid resource business platform to obtain the equipment list and real-time data. If the API request fails, a failure log is printed. If the API request succeeds, the equipment list is parsed and invalid equipment is filtered out. The valid equipment list data is written to the DataList.json file, and local historical data is loaded. 3D equipment icons are generated based on the equipment data. The three status options of equipment defects, hidden dangers, and faults are initialized and stored in the corresponding Map (dictionary) to realize the association between data and the three-dimensional digital twin model. After initialization, the digital twin operation carrier enters the "waiting for front-end instructions" state. When it receives the front-end "getTaiZhang" (get ledger) request, it parses the request parameters and calls the ledger interface to obtain the detailed equipment ledger and update the front-end UI (such as displaying equipment model and maintenance history), thereby completing the digital twin platform construction and real-time data access process.

[0080] Furthermore, such as Figure 2 As shown, step S1 specifically includes:

[0081] Step S1-1: The entire area of ​​the substation is scanned by combining laser point cloud with oblique photography to generate high-density point cloud data. Based on the high-density point cloud data, GIS coordinate registration is performed to unify the point cloud data, image data, and wiring diagram into the same coordinate system. GIS coordinate registration refers to the process of unifying spatial data from different sources and coordinate systems into the same coordinate system through geographic information system technology.

[0082] Steps S1-2: When constructing a 3D digital twin, the 3D entity is meshed, LOD lightweighted, and WebGL loadable processed. Device node encoding is performed, and a unique ID is generated for each device as a data anchor point. LOD lightweighting is a 3D model optimization technique that dynamically adjusts the level of detail of the model according to the viewing distance or display requirements. WebGL loadable processing is a JavaScript API for rendering 3D graphics in a browser. By performing format conversion, compression, texture optimization, and other processing on the 3D model, it can be smoothly loaded and displayed on the Web.

[0083] Steps S1-3: Access the real-time measurement center data through the gateway. The real-time measurement center data includes real-time power, temperature, defects, hidden dangers and historical maintenance work orders. Build an equipment defect database based on defects and historical maintenance work orders.

[0084] Steps S1-4: The digital twin platform interacts with web and mobile terminals, and supports viewing in a visual format on web and mobile terminals.

[0085] Furthermore, such as Figure 3 As shown, step S2 specifically includes:

[0086] Step S2-1: Query and obtain the task ID, power outage period, equipment list, and original text of the three measures and one plan for the maintenance task. The three measures and one plan include organizational measures, technical measures, safety measures, and work plan.

[0087] Step S2-2: Segment the task text of the maintenance task, obtain key information, and generate structured text according to the text format template;

[0088] Step S2-3: Query the device nodes that correspond one-to-one with the device ID in the digital twin platform to form a set of maintenance devices with coordinates;

[0089] Step S2-4: Output maintenance constraints. Maintenance constraints include maintenance equipment set, time axis, procedure template and tool list.

[0090] Furthermore, such as Figure 4 As shown, step S3 specifically includes:

[0091] Step S3-1: Perform remote surveying and mapping operations, and calculate the horizontal distance, vertical distance and spatial distance in real time in the three-dimensional digital twin;

[0092] Specifically, such as Figure 9 As shown, when a user clicks on the first measurement point in the 3D scene, it checks if the measurement point is empty. If the measurement point is initially "empty" on the first click, it proceeds to the "Yes" branch and saves the result. Alternatively, a specific device node can be selected via "Button Device 1" and saved as... If the user clicks on another measurement point in the scene again, at this time... Entering the "No" branch, another measurement point is saved as... Alternatively, a specific device node can be selected via "Button Device 2" and saved as... If the digital twin platform interface has a "Enable Measurement" checkbox, checking it will set a boolean variable to mark the measurement function as activated, ultimately triggering a real-time refresh process, requiring the input of two 3D measurement points. and , The starting point for measurement is at coordinates [0, 1]. , The endpoint of the measurement is located at coordinates [omitted]. Detection and If a measurement point is valid, the system calculates the horizontal, vertical, and spatial distances. If not, it waits for the next click. When a survey line connecting two valid 3D measurement points crosses a live equipment area in 3D space, a minimum safe distance check is automatically triggered to determine if the shortest distance between the survey line and the live equipment meets the minimum safe distance requirements stipulated in the relevant safety regulations for substation operation and maintenance. For example, in a remote survey for the maintenance of a 220kV substation main transformer, if measuring the distance between the top of the main transformer bushing and a certain structure, the survey line needs to cross a live equipment area. In the 220kV busbar area, a verification will be triggered at this time. First, the spatial range of the live equipment in the 3D scene needs to be identified. Then, the shortest straight distance between the measuring line and the boundary of the live equipment is calculated and compared with the preset safety threshold. If the calculated shortest distance is greater than or equal to the safety threshold, it is judged as "safe" and subsequent special analysis can continue. If the shortest distance is less than the safety threshold, it is judged as "unsafe", triggering an alarm and prompting the staff to adjust the position of the measuring point to avoid the safety risk of insufficient distance between the staff, equipment and live parts in subsequent operations, and to ensure the compliance of the overall operation plan.

[0093] The formula for calculating horizontal distance is: ;

[0094] The formula for calculating vertical distance is: ;

[0095] The formula for calculating spatial distance is: ;

[0096] in, Horizontal distance Vertical distance For spatial distance, , , These are two three-dimensional measurement points at... , , Coordinate difference on the axis , , This operation can display the corresponding value based on the checkboxes on the interface of the digital twin platform and synchronously store the calculation results;

[0097] Step S3-2: Perform a topology analysis of the power outage area. The input is the adjacency matrix of the primary wiring diagram. The connected subgraph algorithm is used to calculate the power outage area and the live area, and the compliance of the power outage area is indicated.

[0098] Specifically, such as Figure 10 As shown, using the adjacency matrix of a single wiring diagram as input, the adjacency matrix is... phalanx, This represents the total number of nodes in a primary wiring diagram. A primary wiring diagram is a graph that abstracts equipment in a power grid, such as busbars, circuit breakers, disconnectors, lines, transformers, and loads, into "nodes" and "branches." The adjacency matrix is ​​used to describe the primary wiring diagram. square array Assume there are a total of 100 wires in the wiring diagram. There are 1 node, numbered as For any node ,node If node With nodes When there is a direct electrical connection between the elements, such as a closed branch with a non-infinite impedance, the matrix elements... , representing a node With nodes There is a direct electrical connection between the matrix elements. , representing a node With nodes There are no direct connections between them, and the symmetry of the adjacency matrix depends on the branch direction. If we only care about undirected connections, then... By taking this adjacency matrix as input, we can obtain complete topological information on "who is directly connected to whom";

[0099] A connected subgraph algorithm, combined with the on / off states of equipment switches, is used to calculate the power outage and energized regions. The power outage region is the set of nodes disconnected from all power sources, while the energized region is the set of nodes still connected to power sources. The connected subgraph algorithm is implemented based on Depth-First Search (DFS), Breadth-First Search (BFS), or Union-Find. A connected component is a maximal subgraph formed by all mutually reachable nodes in a single wiring diagram. Whether a node is "reachable" depends on whether the branch is energized, determined by the switch's on / off state. The connected subgraph algorithm can be used with a given graph. Given an optional set of failed elements, calculate the remaining graph after removing the failed elements. Given all connected components, return a list of nodes contained in each connected component and the component number to which each node belongs. Failed elements include failed edges and failed nodes.

[0100] Therefore, this connected subgraph algorithm additionally considers the switch state vector, treating a closed switch as an edge existing and an open switch as an edge removed, thus dividing the region into: the set of nodes still connected to the power source is the energized region, and the set of nodes disconnected from all power sources is the de-energized region. This process defines... The total number of nodes is and the complete set of nodes is . The adjacency matrix of a single-line connection diagram (undirected connection graph) is: The set of power outage nodes is , List the node numbers that have exited the operation; the connected subgraph algorithm will treat them as "powered out".

[0101] The substation primary wiring diagram is divided into multiple interconnected but non-overlapping sub-regions, specifically including: [The text abruptly ends here, likely due to an incomplete sentence or missing information.] The function, with the formula: Based on adjacency matrix Describe the direct electrical connections between nodes, dividing the entire primary wiring diagram into... A connected component, satisfying , , And each exist The induced subgraph is a connected maximal connected subgraph. Represents any first A connected component is a unit that contains at least one node that is not an empty region. The representative puts 1 to All nodes of a connected component are merged, and this merge equals the total set of nodes in the entire one-dimensional wiring diagram. , It represents any two distinct connected components that have no overlapping common nodes and are completely independent of each other;

[0102] Calculate the power outage area: In other words, the power outage area is the union of "all connected components containing at least one power outage node". It is the set of all connected components obtained from the previous calculation using connected components. , This indicates that only the set of nodes experiencing power outages will be selected and retained. "Intersecting connected components, The final power outage area refers to all nodes that must be de-energized during the power outage operation.

[0103] Calculate the charged region: This refers to the remaining part of a wiring diagram after removing the nodes that are de-energized. These nodes are still connected to the power supply and remain energized. Area affected by power outage This is a charged area;

[0104] Through the above operations, such as Figure 10 As shown, each connected component is traversed. If a connected component intersects with a power outage node, the component is assigned to the "power outage area," and components without intersection are assigned to the "energized area." This operation facilitates the identification of which equipment must be de-energized and which equipment remains energized, which is beneficial for subsequent maintenance and safety measures, and ensures the normal operation of certain areas of the power grid. and Used for subsequent visualization rendering Used to output verification results and alarm information to the operation and maintenance terminal;

[0105] Compliance verification of power outage areas is performed using three indicators: internal connectivity, ring network isolation, and scale limitation. Internal connectivity is achieved by generating a power outage area induced subgraph and detecting its connectivity. Ring network isolation is achieved by determining whether the power outage area and the energized area form a loop. Scale limitation is achieved by comparing the number of nodes in the power outage area with a preset threshold. If any one of the verifications fails, an alarm is triggered.

[0106] Specifically, compliance verification is conducted for the power outage area, with key compliance indicators including: internal connectivity ( ), ring network isolation ( ), ring network isolation ( The formula for determining compliance is as follows: Three indicators are used to assess whether the power outage area meets safety and operational requirements. Only when , , When all are TRUE, If any condition is not met, the condition is TRUE; otherwise, it is FALSE and an alarm is triggered. Internal connectivity ( This refers to the property of maintaining electrical connectivity between network nodes consisting of a busbar of the same voltage level and all its directly connected equipment in a primary wiring diagram. The core is determining whether nodes within a power outage area can reach each other through electrical connections. If there is a path between any two nodes consisting only of a closed switch or disconnector, then the busbar area is said to have internal connectivity. Conversely, if the disconnection of a switch or disconnector results in no path between nodes, then internal connectivity is broken, and ring network isolation is compromised. Ring network isolation refers to the ability to divide a ring network into two or more independent sub-networks that are not connected to each other in a ring network structure composed of multiple busbars or multiple power sources by operating switches or disconnectors. Having ring network isolation means that during maintenance or fault handling, the energized parts can be reliably separated from the de-energized parts, ensuring that there is no reverse power supply path in the de-energized area, thereby meeting the requirements for safe operation.

[0107] Determine if internal connectivity exists ( The steps include:

[0108] The determination formula includes: , For induced subgraphs, , This indicates a call to the connectivity detection function. Induced subgraph This is a sub-graph obtained from the original primary wiring diagram, retaining only the nodes in the power outage area and only the original electrical connections between these nodes, while deleting all other nodes and edges. This is the set of edges in the original primary wiring diagram. Edges represent electrical connections between nodes, such as those formed when switches or disconnectors are closed. and It is a power outage area Any two nodes in, This means retaining only edges where both ends are within the power outage area, and then determining the induced subgraph after generating it. Whether a connection is established depends on the following: , That is: any two nodes within the power outage area and There exists a path that is "entirely within the power outage area," making , ,in , They are two adjacent nodes on the path. This represents an undirected edge formed by two adjacent nodes on a path. This represents an undirected edge formed by two adjacent nodes on a path. The edge set is used to examine the edges of each adjacent node along the entire path, ensuring that every edge exists within the edge set of the induced subgraph of the power outage area. In the middle, if any two nodes If a path can be found between all such nodes, then the "internal connectivity of the power outage area" is determined to be TRUE, if at least one pair of nodes exists. If no internal path is found, then the internal connectivity is determined to be "FALSE".

[0109] Determine whether ring network isolation exists ( The steps include:

[0110] The determination formula is: That is, if "a loop exists between the power outage area and the energized area", then FALSE, otherwise TRUE. This is an undirected adjacency matrix of a single wiring diagram. , Represents a node ,node Directly connected; otherwise, 0. , The set of nodes representing the power outage area. , The set of nodes representing the charged region. , The adjacency matrix of the induced subgraph representing the power outage area retains only the following. Internal node connections, , The adjacency matrix of the induced subgraph representing the charged region retains only the values ​​of the induced subgraph. Internal node connections, , This is a cross-regional connection matrix that records the connections between power outage areas and energized areas;

[0111] The condition for the existence of a cycle is: if there exists a positive integer... and , making and Then determine the isolation of the ring network ( If the condition is false, then it is true; otherwise, it is true. Used to determine whether there are nodes within the power outage area. To the node The internal path, Used to determine whether there are slave nodes within a charged region. To the node The internal path, Nodes indicating the power outage area Nodes with charged regions Direct connection, Nodes indicating the power outage area Nodes with charged regions Direct connection: If all four conditions above are met, it means that the power outage area and the energized area form a loop through two cross-boundary edges and their respective internal paths, which does not meet the requirements for ring network isolation.

[0112] Size restrictions ( The determination steps include:

[0113] The determination formula is: , This represents the total number of nodes in the power outage area. The maximum number of nodes (or devices) allowed to be powered off simultaneously is a pre-defined positive integer threshold, set by operation and maintenance specifications, scheduling procedures, or risk assessments. If the total number of nodes experiencing power outages is less than or equal to the maximum number of nodes, then a scale limit is applied. If TRUE is true, then FALSE is true; otherwise, FALSE is true.

[0114] Through the above operations, such as Figure 10 As shown, through internal connectivity ( ), ring network isolation ( ) and size limitations ( Three key indicators are used to strictly determine whether a power outage plan is safe and compliant;

[0115] Step S3-3: Perform a safety measure pre-simulation operation to generate a three-dimensional virtual fence from the electrified boundary and generate a safety measure layout diagram that includes fence coordinates, length, sign position and orientation.

[0116] According to the enclosure of the electrical equipment, respectively , , The minimum and maximum values ​​on the axis, along with the preset safe expansion distance, are used to calculate the two-dimensional expansion corner point, and then the two-dimensional expansion corner point is moved along... Shaft lifting preset height This yields a sequence of three-dimensional virtual fence nodes, generating a closed three-dimensional virtual fence.

[0117] Specifically, such as Figure 11 As shown, the Ancuo pre-simulation algorithm process is as follows: The digital twin platform initializes the UI interface and waits for user interaction. Users can select the device type, such as maintenance equipment or power outage equipment. If maintenance equipment is selected, the maintenance equipment boundary can be generated by highlighting the device. After the user clicks the "Generate Fence" button, the digital twin platform calls the BP_WeiLan algorithm (the preset fence generation algorithm) to automatically generate a closed virtual fence, and supports operations such as "Continue Placement", "Delete All", and "End Placement".

[0118] To generate a 3D virtual fence at a distance from the charged boundary, it is necessary to first calculate the ground, i.e., the 2D corner points, and then along... The axis is lifted to obtain the three-dimensional nodes, and finally the three-dimensional coordinate sequence of the three-dimensional virtual fence nodes is calculated. The steps include: first, given a charged equipment enclosure (AABB), which can be represented as... The enclosure is in , , The minimum and maximum values ​​on the axis, combined with the preset safe outward expansion distance. Calculate the four outer corner points: , , , ,make If it is necessary to close the fence to form a ring, then let And join ,right Each point in ,along Axis lifting and translation , It is a scalar constant. The axis is the upward direction, a unit vector. This yields the final 3D nodes and generates the final node sequence. That is, after the equipment enclosure is expanded outwards at the four corners of the ground, it is then raised uniformly. 3D fence nodes;

[0119] The total length of the three-dimensional virtual fence is calculated by accumulating the Euclidean distance between adjacent nodes using a linear approximation method. The fence is divided into equal parts, and signs are placed at the midpoint of each fence segment. The position and orientation of the signs are determined, and a safety layout diagram containing fence coordinates, length, sign position and orientation is generated.

[0120] Specifically, to generate a safety measure layout diagram that includes fence coordinates, length, sign location, and orientation, it is necessary to first calculate the fence length and segments, and the sign location and orientation. The steps include: the spline length formula is... The spline curve of the fence (composed of a sequence of nodes) The generated data is approximately composed of multiple straight line segments, each with a length equal to the length of adjacent nodes. and Euclidean distance, The total length is the sum of the lengths of all straight line segments. Average score Each section of the fence is [length missing]. Therefore, the first If the center point of a section of guardrail lies on the curve, then the center point of each section is: , , It is a spline curve function of the fence, and then signs are placed at the midpoint of each segment. The midpoint of the segment is The location of the sign is recorded as: , The sign should face the following direction: The orientation of the sign is given by the unit vector of the tangent vector of the fence curve at that point after vertical correction, ensuring that the sign faces outward for easy identification. Based on the above operations, a safety arrangement diagram containing fence coordinates, length, sign position and orientation can be generated.

[0121] Traverse any node in the 3D virtual fence, check whether the distance from the node to the live equipment meets the safety threshold. If it does not meet the threshold, store the node that does not meet the threshold in the alarm set and trigger an alarm.

[0122] Specifically, after the signs are placed, collision and safety checks are performed, traversing the node set. At each point, check whether the distance to the live equipment is ≥ the safety threshold. If all points meet the requirements, the system is safe. If any point does not meet the requirements, an alarm is triggered, and these non-compliant points are added to the alarm set with the message "Insufficient distance between the fence and the live equipment". It is a point The shortest distance to live equipment The safety threshold should be selected based on actual environmental requirements.

[0123] Step S3-4: Perform vehicle path planning, construct a passable network, generate the shortest collision-free path, and verify the turning radius and safety distance. The output is a vehicle attitude sequence with timestamps. The passable network includes the road centerline, height restriction data, and obstacle data.

[0124] Specifically, such as Figure 12 As shown, the vehicle path planning algorithm process is as follows: After starting the system, the user selects the equipment type. If it is "maintenance equipment", the equipment is highlighted and a safety fence is generated. If it is not "maintenance equipment", it is added to the power outage list and the equipment is highlighted. Then, the type of vehicle to participate in the maintenance operation is selected, the corresponding 3D entity and parameters of the vehicle are loaded, and the path of the vehicle in the substation is planned to ensure that it can reach the target equipment location. The vehicle travels according to the path. During this process, the distance between the vehicle and the energized equipment, fences and obstacles is detected in real time. If the safe distance threshold is exceeded, an alarm is triggered and recorded. If it is not exceeded, the vehicle continues to move. During this process, functions such as "play, pause, reset" and "view follow" are supported for easy observation.

[0125] Define vehicle type Path node sequence Road centerline network obstacle collection ,vehicle The minimum turning radius is ,vehicle The safe distance threshold is timestamp sequence ;

[0126] Filter the set of feasible nodes whose node height is greater than or equal to the minimum height of the vehicle and whose distance from the node to the obstacle is greater than or equal to the safe distance threshold of the vehicle. Filter the set of feasible edges whose edge length is greater than or equal to the minimum turning radius of the vehicle and whose edge lines do not intersect with the obstacles. Construct a passable network that includes the road centerline, height restriction data and obstacle data.

[0127] To plan a route, we first need to filter the nodes and edges that vehicles can pass through, and define the feasible node set as follows. This indicates that nodes that meet the criteria are selected. :node height ≥ Vehicles minimum height And nodes Shortest distance to obstacle ≥ vehicle safe distance threshold Set the feasible edge set as This indicates that safe edges between feasible nodes have been selected. The length of the side is greater than or equal to the length of the vehicle. Minimum turning radius Furthermore, the lines connecting the edges do not intersect with any obstacles, thus obtaining a passable network as follows: , For border rights, ;

[0128] Find the shortest collision-free path from the starting point to the ending point in a traversable network. The shortest collision-free path satisfies the minimum turning radius of the vehicle and the safety distance threshold. In the middle, starting from the point and the end point Given the boundary conditions, find the path with the minimum total cost. traverse all from arrive arrive That is, the shortest path;

[0129] Based on the first node's time, timestamps are assigned to path nodes according to vehicle speed, generating a vehicle attitude sequence with timestamps on the trajectory. This vehicle attitude sequence includes node spatial coordinates, vehicle quaternion orientation, and corresponding time. Time is then assigned to each node on the path, forming a timeline, with the first node's time as the reference. Based on this, the process is repeated sequentially, with the current node's time... Equal to the previous node time Add the spatial distance between the two nodes Divide by vehicle speed The formula is: This establishes a continuous and uniform time axis for the entire path, ensuring that each point on the path corresponds to a unique moment, ultimately generating a sequence of vehicle attitudes for the trajectory. , , It is the direction of the computation function, i.e., the trajectory. It is a vehicle At path nodes The mapping on can generate a series of triples. , , representing the first of the trajectory 1 node The vehicle is in the Spatial coordinates of each node For the vehicle in the The orientation of the quaternions of each node, For the vehicle to arrive At the corresponding moment of the location, this operation combines the path node with the time axis to form the vehicle's running trajectory, which can ensure that the vehicle's position and direction at any time can be uniquely determined, making it convenient to rehearse vehicle driving, generate operation instructions and other scenarios.

[0130] Furthermore, such as Figure 5As shown, step S4 specifically includes:

[0131] Step S4-1: Perform real-time layer overlay operation in any 3D scene displayed on the digital twin platform. The layers include survey map, power outage area map, safety measure layout map, and vehicle route map.

[0132] Step S4-2: Perform an environment overlay operation on the same 3D scene after the layers are overlaid in real time. The environment includes real-time weather, maintenance distribution map and risk level.

[0133] Step S4-3: Generate a maintenance operation plan through the digital twin platform. The maintenance operation plan includes a structured "three measures and one plan", work order, operation order, 3D scene screenshot, vector DWG diagram and executable browsing file.

[0134] Furthermore, such as Figure 6 As shown, step S5 specifically includes:

[0135] Step S5-1: Transmit on-site maintenance data through inspection equipment. Inspection equipment includes mobile terminals, drones, and inspection robots. On-site maintenance data includes GPS tracks, photos, and defect elimination results.

[0136] Step S5-2: Map the GPS track and photos one by one to the digital twin platform. The digital twin platform will automatically refresh the equipment defect database and ledger status.

[0137] Step S5-3: Compare the actual maintenance data with the maintenance operation plan, calculate the deviation index between the two, and adjust the path planning weight and safety threshold of the subsequent maintenance operation plan.

[0138] Specifically, this method creates a digital twin platform that is synchronized with the physical substation in real time and mapped with high precision through the above steps. After receiving maintenance tasks, the platform can automatically parse the tasks and perform four special analyses on the maintenance equipment set in the twin environment: remote survey, power outage analysis, safety measure simulation, and path planning. After integrating the results and superimposing environmental factors, a complete maintenance operation plan including "three measures and one plan", work order, operation order and 3D visualization guide can be generated with one click. This allows the maintenance operation plan to be compiled based on the real-time on-site status, which solves the problem of static display and disconnect from the site of traditional 3D visualization system, improves the accuracy of the plan, and after the operation is completed, the on-site data can be sent back to the digital twin platform for comparison and the deviation index between the actual situation and the plan can be calculated to optimize the generation parameters of subsequent operation plans. It has good application prospects.

[0139] Example 1

[0140] Taking the annual maintenance of the main transformer and grounding switch of a 220kV substation as an example:

[0141] Step S1: Use a laser scanner to scan the entire 220kV substation along the roads and platforms to generate high-density point cloud data. Collect oblique photographic images of the entire substation using a drone. Retrieve the substation's primary wiring diagram in DWG / SVG format (for web display). This includes the electrical connections of the main transformer, grounding switch, associated busbars, circuit breakers, and other equipment. Perform GIS coordinate registration based on the high-density point cloud data to unify the point cloud data, image data, and wiring diagram into the same coordinate system. When constructing a 3D digital twin, perform point cloud filtering and noise reduction on the 3D entity. Create equipment nodes for the main transformer and grounding switch in a 3D engine (such as Unity). Name the main transformer node "Transformer". The _220_T1 and grounding switch nodes are named ES_220_Bus1_Q1. Ensure smooth loading on both web and mobile devices. Access real-time power, temperature, and other information from the real-time measurement center via the gateway. Example fields obtained: {"deviceId":"T1", "P": 168.4", "Q": 52.3", "T_top_oil": 68.2", "timestamp": "2025-08-28T10:00:00Z"}. Publish via WebGL on the web, supporting visual viewing of the framework on both web and mobile devices. It can access the equipment defect database, recording overheating defects in the grounding switch and the emergency status level. Example fields from the equipment defect database are as follows:

[0142] Step S2: Access the power grid resource service platform, query and obtain the task ID, outage period, equipment list, and original text of the "Three Measures and One Plan" for the maintenance task, export the maintenance plan, and obtain "Three Measures and One Plan.pdf". Perform a segmentation operation on the task text of the maintenance task, parse "Three Measures and One Plan.pdf" (refer to GB26860-2011), and obtain the structured text:

[0143] Organizational Measures = "Work Leader: Zhang San (Qualification Certificate No.: DL-G-2023-1025)"

[0144] Team: Substation Maintenance Team 2 (7 people: 2 primary maintenance workers, 2 testing workers, 2 secondary maintenance workers, 1 safety inspector)

[0145] Technical measures = "Interval switching to maintenance, grounding switch closed"

[0146] Operating procedures: 220kV #1 main transformer 2201 switch → cold standby → maintenance; close 2201-17 grounding switch (ES_220_Bus1_Q1)

[0147] Safety measure = "The distance between the fence and the live parts on the busbar side is ≥3.5m"

[0148] In the digital twin platform, query the device nodes that correspond one-to-one with the device IDs to form a set of maintenance equipment with coordinates. Output the maintenance constraints, which include the maintenance equipment set, timeline, procedure template, and tool list. Example:

[0149] Generate constraints:

[0150] Device = "Main Transformer ID, Switch ID"

[0151] Time window = "8:00, 18:00"

[0152] Safe distance = 3.5m

[0153] Tools = "SF6 recycling truck, aerial work platform truck"

[0154] Step S3: Conduct a special analysis on the maintenance equipment set in the digital twin platform and provide the special analysis results. The special analysis includes remote surveying and mapping, power outage area topology analysis, safety measure simulation and vehicle route planning.

[0155] Remote surveying and mapping operation: The operator clicks on the top bushing of the main transformer on the digital twin platform to obtain the coordinates of the top of the main transformer bushing. According to the formula, the horizontal span is calculated in real time to be 7m, the vertical height difference is 9m, and the shortest distance to the 220kV busbar is 5.11m (if the distance is greater than 3.5m, it passes).

[0156] Power outage area topology analysis: Input the primary wiring diagram adjacency matrix, calculate the outage nodes and energized nodes, mark the energized busbars in red and the outage area in green in the 3D scene, and determine the outage or energized nodes using NetworkX 3.1 DFS. Assuming the node numbers include: 0-220kV #1 main transformer, 1-2201 circuit breaker, 2-2201-17 grounding switch, 3-220kV I busbar...37-110kV line 2, construct the graph from the adjacency matrix A, and use the 220kV busbar (node ​​3) as the source to perform DFS to obtain all reachable nodes;

[0157] Ancuo Rehearsal: Multiple Sections of Live Busbar Composed of a three-dimensional point array of 220kV busbars, using CGCS2000 coordinates in meters, with a safety threshold of [missing information]. Equipment outline : The convex hull or OBB of the main transformer, disconnector, and structure;

[0158] Specifically, the digital twin platform automatically generates fence nodes, calculates their lengths, allows dragging and adjusting individual nodes, recalculates the nearest point distance in real time, and determines safety. It then places "High Voltage Danger" signs in the scene and records their positions. Specifically, it generates a simple polygon outside the power outage area boundary, satisfying the condition that "distance from any live object is ≥..." "For multiple sections of energized busbars" Create a two-dimensional buffer to obtain the polygon of the safe zone. ,right Further expansion, safe expansion distance The initial fence centerline is obtained and simplified using the Douglas-Peucker method. With a threshold of 0.3m, the number of nodes is reduced, and the simplified point list is used as the initial fence node list. The interface is closed at both ends, and uses the WebGL 3D rendering library Three.js for front-end interaction (drag and drop). Each movement only modifies one node. The remaining nodes remain stationary; input the node that was dragged. Multiple sections of live busbar The nearest point distance is calculated, and then real-time nearest point distance recalculation is performed. Then, equidistant sampling is conducted along the fence centerline with a step size of 20m to obtain candidate points. For each candidate point Offset 0.5m outwards in the direction of the outward normal, and we get: Distance from any charged body If so, offset 0.5m in the inward normal direction. This yields the final list of sign coordinates. ,in The normal direction angle;

[0159] Vehicle route planning: Based on the passable network nodes within the substation, calculate the vehicle route: gate → main transformer road → one side of the disconnector. Calculate whether the distance and turning radius meet the requirements. Output the vehicle attitude sequence and timestamp. Input vehicle parameters include: vehicle type, external dimensions, wheelbase, minimum turning radius, design speed, etc. Passable network nodes include the main road centerline, passing zones, and side roads beside equipment, with a width ≥ 3.5m. Use QGIS (Geographic Information System) software to convert the route into a polyline, breaking it every 5m to obtain the node list. Construct a topology graph , For all road nodes, gate nodes, and switch operation points, Target For a straight line segment between two nodes, the following conditions must be met: Euclidean distance ≤ 30m, the entire segment is within a passable surface, the minimum width within the segment is ≥ vehicle width + 2 × 0.3m safety distance, and the maximum slope within the segment is ≤ 8%. The shortest path search with turning radius constraints can use the A-withDubins heuristic algorithm.

[0160] Step S4: Output structured data for the four specialized analyses performed in Step S3. The remote survey module outputs distance data (horizontal, vertical, and spatial distances) and safety verification results. The power outage analysis module outputs the power outage area node set, the energized area node set, and compliance verification results. The safety measure simulation module outputs fence coordinates, length, sign position and orientation, and safety verification results. The vehicle path module outputs the path node sequence, timestamp, and vehicle posture sequence. Overlay the above analysis results onto the 3D scene in real time as layers: survey map (distance annotation), power outage range map (… The system includes red / green zones, safety measures layout diagrams (fences and signs), and vehicle route maps (path trajectories). Layers are overlaid in real-time within any 3D scene displayed on the digital twin platform. These layers include survey scales, power outage red zones, green fences, and blue driving routes. Environmental overlay operations are then performed on the same 3D scene after the layers are overlaid in real-time to generate a maintenance work plan. This plan includes DWG drawings and PNG screenshots exported from documents such as "Three Measures and One Plan.doc" and "Work Order.pdf". The digital twin platform automatically records the version number and provides a QR code for on-site scanning and viewing.

[0161] Step S5: After the maintenance work is completed, the drone takes pictures of the switch temperature dropping to a normal level, records the GPS trajectory file, and updates the equipment defect database status by taking pictures of the fence and tool placement with a mobile phone. The switch overheating defect is "eliminated" and the elimination time is recorded to ensure that the equipment ledger is consistent with the actual on-site status. The actual maintenance data is compared with the maintenance work plan, and deviation indicators such as actual driving time and planned driving time are recorded for adjusting the subsequent plan. For example, the automatic comparison of planned driving time and actual driving time found that the actual time was shortened by 5 minutes. After path backtracking analysis, it was determined that the speed limit setting of the engineering vehicle was too low. The platform can automatically adjust the vehicle speed limit, and it is expected that the corresponding driving time can be shortened in the next operation, so as to achieve continuous optimization of model parameters.

[0162] Specifically, this embodiment can significantly reduce the time required to generate outage range, safety barriers, and vehicle paths through the substation operation plan preparation method, control the errors in outage boundaries, energized parts, and barrier layout, enable real-time collision detection of vehicle paths, and provide immersive 3D simulation for operators, reducing human-caused accidents such as misoperation and accidental entry into gaps. Real-time overlay of weather, load, and defects can identify high-risk periods and equipment in advance, shorten the average outage time for a single maintenance, avoid repetitive modeling and paper documents, and reduce operation and maintenance costs.

[0163] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A method for compiling substation operation plans based on a three-dimensional model, characterized in that, Specifically, the following steps are included: Step S1: Collect and fuse multi-source data from the substation, construct a three-dimensional digital twin, and connect it to real-time data to form a digital twin platform that supports multi-terminal interaction; Step S2: Query the maintenance task and parse the key information in the maintenance task, extract the key information and perform text structuring processing, match the corresponding device node in the digital twin platform through the device ID to form a maintenance equipment set, and output the maintenance constraints. Step S3: Conduct a special analysis on the set of maintenance equipment in the digital twin platform and provide the special analysis results. The special analysis includes remote surveying and mapping, power outage area topology analysis, safety measure simulation and vehicle route planning. Step S4: Integrate the special analysis results, overlay environmental factors, display the maintenance scenario in the digital twin platform, and generate and output the maintenance operation plan; Step S5: After the maintenance work is completed, transmit the on-site maintenance data back and update the data on the digital twin platform; The digital twin platform comprises a perception layer, a twin layer, an algorithm layer, and an interaction layer. The perception layer consists of a lidar module and an oblique photography module, used to collect raw data from the substation. The twin layer includes an equipment node encoding module, a time series caching module, and a real-time data mapping module, used to assign a unique identifier to each device through equipment node encoding and to associate real-time equipment operation data and environmental data with the three-dimensional digital twin through real-time data mapping. The algorithm layer includes a remote survey module, a power outage analysis module, a safety measure pre-simulation module, a vehicle routing module, and a deviation calculation module, used to conduct specialized analysis on the set of maintenance equipment. The interaction layer includes a three-dimensional roaming module, a multi-layer overlay module, a solution packaging module, and a version evolution module, used to receive and output data and instructions and provide operational functions to users. The digital twin platform is equipped with a connection port for data access operations from meteorological APIs, power grid resource business platforms, and defect databases. Step S3 specifically includes: Step S3-1: Perform remote surveying and mapping operations, and calculate the horizontal distance, vertical distance and spatial distance in real time in the three-dimensional digital twin; Step S3-2: Perform a topology analysis of the power outage area. The input is the adjacency matrix of the primary wiring diagram. The connected subgraph algorithm is used to calculate the power outage area and the live area, and the compliance of the power outage area is indicated. Step S3-3: Perform a safety measure pre-simulation operation to generate a three-dimensional virtual fence from the electrified boundary and generate a safety measure layout diagram that includes fence coordinates, length, sign position and orientation. Step S3-4: Perform vehicle path planning, construct a passable network, plan and generate the shortest collision-free path, and verify the turning radius and safety distance. The output is a vehicle attitude sequence with timestamps. The passable network includes road centerline, height restriction data and obstacle data. Step S3-2 specifically includes: The adjacency matrix of the first-order wiring diagram is used as input, and the adjacency matrix is... phalanx, Given the total number of nodes in a wiring diagram, if the matrix elements... , representing a node ,node There is a direct electrical connection between the matrix elements. , representing a node With nodes There is no direct connection between them; The connected subgraph algorithm is used in conjunction with the on / off state of the equipment switch to calculate the power outage area and the energized area. The power outage area is the set of nodes that have lost connection with all power sources, and the energized area is the set of nodes that are still connected to the power sources. Compliance verification of power outage areas is performed using three indicators: internal connectivity, ring network isolation, and scale limitation. Internal connectivity is achieved by generating a power outage area induced subgraph and detecting its connectivity. Ring network isolation is achieved by determining whether the power outage area and the energized area form a loop. Scale limitation is achieved by comparing the number of nodes in the power outage area with a preset threshold. If any one of the verifications fails, an alarm is triggered.

2. The method for compiling substation operation plans based on a three-dimensional model according to claim 1, characterized in that, Step S1 specifically includes: Step S1-1: The entire area of ​​the substation is scanned by combining laser point cloud with oblique photography to generate high-density point cloud data. Based on the high-density point cloud data, GIS coordinate registration is performed to unify the point cloud data, image data, and wiring diagram into the same coordinate system. Step S1-2: When constructing a 3D digital twin, the 3D entity is meshed, LOD lightweighted, and WebGL loadable processed, and device node encoding is performed to generate a unique ID for each device as a data anchor point. Steps S1-3: Access the real-time measurement center data through the gateway. The real-time measurement center data includes real-time power, temperature, defects, hidden dangers, and historical maintenance work orders. Construct an equipment defect database based on the defects and historical maintenance work orders. Steps S1-4: The digital twin platform interacts with web and mobile terminals and supports viewing in a visual format on web and mobile terminals.

3. The method for compiling a substation operation plan based on a three-dimensional model according to claim 1, characterized in that, Step S2 specifically includes: Step S2-1: Query and obtain the task ID, power outage period, equipment list, and original text of the three measures and one plan for the maintenance task. The three measures and one plan include organizational measures, technical measures, safety measures, and work plan. Step S2-2: Segment the task text of the maintenance task, obtain key information, and generate structured text according to the text format template; Step S2-3: Query the device nodes that correspond one-to-one with the device ID in the digital twin platform to form a set of maintenance devices with coordinates; Step S2-4: Output maintenance constraints, which include a set of maintenance equipment, a timeline, a procedure template, and a list of tools and equipment.

4. The method for compiling a substation operation plan based on a three-dimensional model according to claim 1, characterized in that, Step S3-1 specifically includes: Input two 3D measurement points and , The starting point for measurement is at coordinates [0, 1]. , The endpoint of the measurement is located at coordinates [omitted]. Detection and If a measurement point is valid, the horizontal distance, vertical distance, and spatial distance are calculated. When the survey line connecting two valid three-dimensional measurement points crosses the area of ​​live equipment in three-dimensional space, the shortest safe distance check is automatically triggered. The formula for calculating horizontal distance is: ; The formula for calculating vertical distance is: ; The formula for calculating spatial distance is: ; in, Horizontal distance Vertical distance For spatial distance, These are two three-dimensional measurement points at... Coordinate difference on the axis , , .

5. The method for compiling a substation operation plan based on a three-dimensional model according to claim 1, characterized in that, Step S3-3 specifically includes: According to the enclosure of the electrical equipment, respectively The minimum and maximum values ​​on the axis, along with the preset safe expansion distance, are used to calculate the two-dimensional expansion corner point, and then the two-dimensional expansion corner point is moved along... Shaft lifting preset height This yields a sequence of three-dimensional virtual fence nodes, generating a closed three-dimensional virtual fence. The total length of the three-dimensional virtual fence is calculated by accumulating the Euclidean distance between adjacent nodes using a linear approximation method. The fence is divided into equal parts, and signs are placed at the midpoint of each fence segment. The position and orientation of the signs are determined, and a safety layout diagram containing fence coordinates, length, sign position and orientation is generated. Traverse any node in the 3D virtual fence, check whether the distance from the node to the energized equipment meets the safety threshold. If it does not meet the threshold, store the node in the alarm set and trigger an alarm.

6. The method for compiling a substation operation plan based on a three-dimensional model according to claim 1, characterized in that, Step S3-4 specifically includes: Filter the set of feasible nodes whose node height is greater than or equal to the minimum height of the vehicle and whose distance from the node to the obstacle is greater than or equal to the safe distance threshold of the vehicle. Filter the set of feasible edges whose edge length is greater than or equal to the minimum turning radius of the vehicle and whose edge lines do not intersect with the obstacles. Construct a passable network that includes the road centerline, height restriction data and obstacle data. Find the shortest collision-free path from the starting point to the ending point in a traversable network, wherein the shortest collision-free path satisfies the minimum turning radius of the vehicle and the safety distance threshold. Based on the first node time, timestamps are assigned to path nodes according to vehicle speed to generate a vehicle attitude sequence of running trajectory with timestamps. The vehicle attitude sequence includes node spatial coordinates, vehicle quaternion orientation and corresponding time.

7. The method for compiling substation operation plans based on a three-dimensional model according to claim 1, characterized in that, Step S4 specifically includes: Step S4-1: Perform real-time layer overlay operation in any three-dimensional scene displayed on the digital twin platform. The layers include survey map, power outage area map, safety measure layout map, and vehicle route map. Step S4-2: Perform an environment overlay operation on the same 3D scene after real-time layer overlay. The environment includes real-time weather, maintenance distribution map, and risk level. Step S4-3: Generate a maintenance operation plan through the digital twin platform. The maintenance operation plan includes a structured three-measure plan, a work order, an operation order, a 3D scene screenshot, a vector DWG image, and an executable browsing file.

8. The method for compiling a substation operation plan based on a three-dimensional model according to claim 1, characterized in that, Step S5 specifically includes: Step S5-1: Transmit on-site maintenance data through inspection equipment, including mobile terminals, drones and inspection robots, and the on-site maintenance data includes GPS trajectory, photos and defect elimination results. Step S5-2: Map the GPS track and photos one by one to the digital twin platform. The digital twin platform automatically refreshes the device defect database and ledger status. Step S5-3: Compare the actual maintenance data with the maintenance operation plan, calculate the deviation index between the two, and adjust the path planning weight and safety threshold of the subsequent maintenance operation plan.

Images