A power transmission and transformation engineering management and control method and system based on digital twinning

Abstract

The application provides a power transmission and transformation engineering management and control method and system based on digital twinning, relates to the technical field of management and control, acquires real-time multi-source heterogeneous data of a power transmission and transformation engineering; based on an engineering multi-dimensional digital twinning model, time-space alignment and semantic fusion are performed on the real-time multi-source heterogeneous data to generate a real-time virtual mirror; according to the real-time virtual mirror, power transmission and transformation process simulation deduction is driven to obtain simulation deduction results; based on a management and control rule library and the simulation deduction results, multi-dimensional engineering management and control instructions are generated, the engineering management and control instructions are issued to an execution terminal of the power transmission and transformation engineering, and the real-time running state of the power transmission and transformation engineering is adjusted. The application can construct a digital twinning body with causal perception capability and drive double-channel parallel simulation deduction, and the accuracy of abnormal prediction of the power transmission and transformation engineering and the synergy of multi-dimensional regulation and control are improved.

Description

Technical Field

[0001] This application relates to the field of control technology, and more specifically, to a control method and system for power transmission and transformation projects based on digital twins. Background Technology

[0002] Existing power transmission and transformation project management technologies largely rely on a combination of data acquisition and monitoring control systems, equipment condition monitoring systems, and statistical process control. This involves real-time acquisition and threshold comparison of operating parameters of power transmission and transformation equipment to monitor and alarm on the project status and equipment health. Building upon this, some technologies introduce digital twin models. By constructing a three-dimensional geometric model of the power transmission and transformation project and integrating real-time data, the system visually reproduces the equipment status and environmental information at the project site, assisting managers in experience-based decision-making and post-event retrospective analysis. In terms of simulation and analysis, existing technologies typically employ methods based on power flow calculations or electromechanical transient simulations. Based on a pre-defined network topology and operating mode, they simulate and analyze the power grid's operating status to assess system stability and power supply reliability.

[0003] However, while existing power transmission and transformation engineering management technologies can detect single-point anomalies, they struggle to reveal the causal propagation paths of faults within complex power networks. For example, how a single equipment failure can cascade through electrical connections and protection coordination, leading to regional voltage instability or widespread power outages. Furthermore, current technologies are insufficient for proactive prediction before incidents escalate; most management measures remain reactive and focused on localized, single-point responses, failing to achieve multi-dimensional, coordinated control encompassing equipment status regulation, grid topology adjustments, and protection strategy optimization. Therefore, constructing a digital twin with causal awareness capabilities and driving dual-channel parallel simulation to improve the accuracy of anomaly prediction and the synergy of multi-dimensional control in power transmission and transformation engineering remains a challenge for the industry. Summary of the Invention

[0004] This application provides a digital twin-based method and system for the management and control of power transmission and transformation projects. It can construct a digital twin with causal perception capabilities and drive dual-channel parallel simulation and deduction, thereby improving the accuracy of anomaly prediction and the synergy of multi-dimensional control in power transmission and transformation projects.

[0005] In a first aspect, this application provides a management and control method for power transmission and transformation projects based on digital twins, the management and control method comprising the following steps: Acquire real-time multi-source heterogeneous data from power transmission and transformation projects; Based on a pre-built multidimensional digital twin model of the project, the real-time multi-source heterogeneous data is spatiotemporally aligned and semantically fused to generate a real-time virtual image of the power transmission and transformation project. Based on the real-time virtual image, the power transmission and transformation process is simulated and deduced to identify abnormal events and potential weak links in the power transmission and transformation process, and to obtain the simulation and deduction results. Based on the preset control rule base and the simulation results, multi-dimensional engineering control instructions are generated and sent to the execution terminal of the power transmission and transformation project to adjust the real-time operation status of the power transmission and transformation project.

[0006] In this embodiment, the multi-source heterogeneous data includes: operating status data of power transmission and transformation equipment, power grid topology and power flow data, and engineering environment data.

[0007] In this embodiment, based on a pre-built multi-dimensional digital twin model of the project, spatiotemporal alignment and semantic fusion are performed on the real-time multi-source heterogeneous data to generate a real-time virtual image of the power transmission and transformation project. Specifically, this includes: A pre-constructed multidimensional digital twin model of an engineering project is obtained. The multidimensional digital twin model of an engineering project includes a geometric structure layer, a physical attribute layer, a behavioral rule layer, and a semantic association layer. The layers are associated with each other through a preset mapping relationship. The real-time multi-source heterogeneous data is registered in time and transformed in space to align data streams with different sampling frequencies to a unified time reference, and the spatial coordinate system of each data point is transformed to the global coordinate system of the engineering multidimensional digital twin model. The spatiotemporally aligned multi-source heterogeneous data is injected into the corresponding level of the engineering multidimensional digital twin model to drive the pose and appearance state update of the geometric structure layer, the field quantity distribution update of the physical attribute layer, and the state transition and dynamic behavior response of the behavior rule layer. In the semantic association layer, the updated entity nodes at each level are associated and bound according to the predefined semantic relationship network to generate a real-time operation status diagram with causal association weights. The real-time operation status diagram is a real-time virtual image of the power transmission and transformation project.

[0008] In this embodiment, generating a real-time operational status map with causal correlation weights specifically includes: Assign initial causal association weights to the association edges between entity nodes in the semantic relationship network; When the real-time state value of a certain entity node deviates from its steady-state baseline, it is marked as a disturbance source node; Starting from the disturbance source node, the disturbance spreads outward along the association edges of the semantic relationship network, and the disturbance degree value of each downstream association node is determined based on the causal association weight of each association edge. The disturbance values ​​of each node are superimposed along the propagation path to form a causal disturbance distribution map, which in turn generates a real-time operational status map with causal correlation weights.

[0009] In this embodiment, the initial causal association weight is preset by expert experience based on the electrical connection relationship and protection coordination logic of the power transmission and transformation project.

[0010] In this embodiment, based on the real-time virtual image, the power transmission and transformation process is simulated and deduced to identify abnormal events and potential weak points in the power transmission and transformation process. The simulation and deduction results specifically include: Extract a snapshot of the current running state from the real-time virtual image; Using the current running state snapshot as the initial condition, the dual-channel parallel simulation engine is driven to perform the deduction in the behavior rule layer of the multidimensional digital twin model of the project; The dual-channel parallel simulation engine includes: a first channel that runs at a rate proportional to the physical clock to perform device-level state evolution simulations; and a second channel that runs at a rate faster than the physical clock to perform system-level dynamic simulations of power flow and stability. An abnormal event list is extracted from the simulation results of the first channel, and a weak link warning list is extracted from the simulation results of the second channel. The two are then combined as the simulation results.

[0011] In this embodiment, the device-level refined state evolution deduction specifically includes: For each key power transmission and transformation equipment, extract current characteristic parameters that characterize the health status of the equipment; The current feature parameters are input into a pre-established equipment state prediction model to obtain the equipment state evolution trend over a future period of time; During the evolution simulation, the evolution trend is corrected based on the upstream disturbances reflected in the real-time operational status diagram. When the state evolution trend reaches the alarm threshold within a preset look-ahead time window, an abnormal event record is generated.

[0012] In this embodiment, the generation of multi-dimensional engineering control instructions based on a preset control rule base and the simulation results specifically includes: Extract the abnormal equipment identifier, abnormal type and first boundary crossing time from the abnormal event list from the simulation results, and extract the weak node identifier, instability nucleation time and root cause node information from the weak link early warning list; The extracted information is matched with a preset control rule base at multiple levels. The control rule base includes: equipment control rule cluster, power grid topology adjustment rule cluster, and protection strategy adjustment rule cluster. Based on the rule matching results, a set of coordinated control instructions is constructed; The coordinated control instruction set is converted into engineering control instructions adapted to the execution terminal of physical power transmission and transformation projects.

[0013] In this embodiment, the coordinated control instruction set includes: equipment operating parameter control strategy, power grid topology scheduling strategy, and protection control strategy.

[0014] Secondly, this application provides a digital twin-based power transmission and transformation project management and control system for executing a digital twin-based power transmission and transformation project management and control method, the management and control system comprising: The data acquisition module is used to acquire real-time multi-source heterogeneous data from power transmission and transformation projects. The image generation module is used to perform spatiotemporal alignment and semantic fusion on the real-time multi-source heterogeneous data based on the pre-built multi-dimensional digital twin model of the project, and generate a real-time virtual image of the power transmission and transformation project. The simulation and deduction module is used to drive the simulation and deduction of the power transmission and transformation process based on the real-time virtual image, identify abnormal events and potential weak links in the power transmission and transformation process, and obtain simulation and deduction results; The instruction generation module is used to generate multi-dimensional engineering control instructions based on a preset control rule base and the simulation results, and to send the engineering control instructions to the execution terminal of the power transmission and transformation project to adjust the real-time operating status of the power transmission and transformation project.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The system acquires real-time multi-source heterogeneous data of the power transmission and transformation project; based on a pre-built multi-dimensional digital twin model of the project, it performs spatiotemporal alignment and semantic fusion on the real-time multi-source heterogeneous data to generate a real-time virtual image of the power transmission and transformation project; based on the real-time virtual image, it drives the simulation and deduction of the power transmission and transformation process, identifies abnormal events and potential weak links in the power transmission and transformation process, and obtains simulation and deduction results; based on a preset control rule base and the simulation and deduction results, it generates multi-dimensional engineering control instructions and sends the engineering control instructions to the execution terminal of the power transmission and transformation project to adjust the real-time operating status of the power transmission and transformation project.

[0016] Therefore, this application firstly acquires real-time multi-source heterogeneous data of power transmission and transformation projects by deploying a multimodal sensing network. This reduces the amount of data transmission while preserving key characteristics of equipment status, providing a data foundation for driving the subsequent digital twin model. Secondly, by injecting multi-source heterogeneous data into the digital twin model under a unified spatiotemporal reference and assigning relational edges between entity nodes in the semantic association layer, the generated real-time virtual image can reproduce the explicit operating state of the power transmission and transformation project and reflect the causal transmission path of disturbances in the electrical network, providing a virtual mapping foundation with causal perception capabilities for subsequent simulation and joint control decisions. Thirdly, by using the snapshot of the operating state and the causal disturbance distribution in the real-time virtual image as initial conditions, a dual-channel parallel simulation engine is driven to achieve forward-looking identification of equipment anomalies, providing early warning information with both time margin and root cause orientation for control decisions. Finally, by performing multi-level matching between simulation results and the control rule base, precise closed-loop response to disturbance events and systematic coordinated control of the power transmission and transformation project are achieved.

[0017] In summary, the technical solution adopted in this application can construct a digital twin with causal perception capabilities and drive dual-channel parallel simulation and deduction, thereby improving the accuracy of anomaly prediction and the synergy of multi-dimensional control in power transmission and transformation projects. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only for this embodiment of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is an exemplary flowchart of a digital twin-based power transmission and transformation engineering management and control method provided in this application; Figure 2 This is a schematic diagram of the overall architecture of the power transmission and transformation engineering control system based on digital twin provided in this application; Figure 3 This is a module structure diagram of a power transmission and transformation engineering control system based on digital twins, provided in this application. Detailed Implementation

[0020] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of this application.

[0021] This application provides a digital twin-based method and system for the management and control of power transmission and transformation projects. Its core is to acquire real-time multi-source heterogeneous data of the power transmission and transformation project; based on a pre-constructed multi-dimensional digital twin model of the project, perform spatiotemporal alignment and semantic fusion on the real-time multi-source heterogeneous data to generate a real-time virtual image of the power transmission and transformation project; based on the real-time virtual image, drive the simulation and deduction of the power transmission and transformation process, identify abnormal events and potential weaknesses in the process, and obtain simulation and deduction results; based on a preset management and control rule base and the simulation and deduction results, generate multi-dimensional project management and control instructions, and send these instructions to the execution terminal of the power transmission and transformation project to adjust the real-time operating status of the project.

[0022] Example 1: To better understand the above technical solution, the following will provide a detailed description of the technical solution in conjunction with the accompanying drawings and specific implementation methods. (Refer to...) Figure 1 As shown in the figure, this is an exemplary flowchart of a digital twin-based power transmission and transformation engineering management and control method according to this embodiment of the present application. The management and control method includes the following steps: In step S1, real-time multi-source heterogeneous data of the power transmission and transformation project are acquired.

[0023] In practical implementation, for equipment operating status data, vibration sensors are installed on the transformer tank walls, current transformers and voltage transformers are installed at the bushings, oil temperature sensors are installed on the top, and online oil chromatography monitoring devices are installed in the oil circulation pipelines; vibration sensors and coil current sensors are installed on the circuit breaker operating mechanisms, and SF6 gas density sensors are installed in the arc-extinguishing chambers; angle sensors and torque sensors are installed on the disconnector operating mechanisms; and ultrasonic and UHF partial discharge sensors are installed on the GIS equipment casing. For power grid topology and power flow data, the opening and closing status of circuit breakers and disconnectors, bus voltage, line power, system frequency, and other data are read in real time from the SCADA system via the IEC 61850 interface. For engineering environmental data, sensors for temperature and humidity, wind speed and direction, rainfall, and particulate matter concentration are deployed in various areas of the substation and at transmission line towers, and are aggregated and uploaded via a wireless network. This multi-source heterogeneous data includes: operating status data of transmission and transformation equipment, power grid topology and power flow data, and engineering environmental data.

[0024] In step S2, based on the pre-built multidimensional digital twin model of the project, the real-time multi-source heterogeneous data is spatiotemporally aligned and semantically fused to generate a real-time virtual image of the power transmission and transformation project.

[0025] In this embodiment, based on a pre-built multi-dimensional digital twin model of the project, spatiotemporal alignment and semantic fusion are performed on the real-time multi-source heterogeneous data to generate a real-time virtual image of the power transmission and transformation project. Specifically, this can be achieved through the following steps: A pre-constructed multidimensional digital twin model of an engineering project is obtained. The multidimensional digital twin model of an engineering project includes a geometric structure layer, a physical attribute layer, a behavioral rule layer, and a semantic association layer. The layers are associated with each other through a preset mapping relationship. The real-time multi-source heterogeneous data is registered in time and transformed in space to align data streams with different sampling frequencies to a unified time reference, and the spatial coordinate system of each data point is transformed to the global coordinate system of the engineering multidimensional digital twin model. The spatiotemporally aligned multi-source heterogeneous data is injected into the corresponding level of the engineering multidimensional digital twin model to drive the pose and appearance state update of the geometric structure layer, the field quantity distribution update of the physical attribute layer, and the state transition and dynamic behavior response of the behavior rule layer. In the semantic association layer, the updated entity nodes at each level are associated and bound according to the predefined semantic relationship network to generate a real-time operation status diagram with causal association weights. The real-time operation status diagram is a real-time virtual image of the power transmission and transformation project.

[0026] In practical implementation, firstly, a pre-built multi-dimensional digital twin model of the project can be obtained. That is, a pre-built multi-dimensional digital twin model of the project can be loaded from the model library of the digital twin platform. The multi-dimensional digital twin model of the project is not a single three-dimensional geometric model, but a composite model with a four-layer architecture. The first layer is the geometric structure layer, which contains the three-dimensional geometric model of all physical elements within the substation, such as: main transformer, circuit breaker, disconnector, gas-insulated switchgear, instrument transformer, surge arrester, busbar, insulator, tower, transmission line, etc. All models have accurate spatial coordinates and dimensional information, and their coordinate system is based on the geodetic coordinate system in the substation design drawings. Kinematic constraints have been established for each moving part. The second layer is the physical attribute layer, attached to the geometric structure layer, which contains the physical characteristic description of each physical element. For transformer equipment, the physical attribute layer includes parameters such as rated capacity, rated voltage ratio, short-circuit impedance, no-load loss, load loss, oil heat capacity, heat dissipation coefficient, and winding material. For circuit breaker equipment, the parameters include: rated voltage, rated current, rated breaking current, closing time, opening time, operating mechanism type, and mechanical characteristic parameters. For transmission lines, the parameters include: resistance per unit length, reactance, susceptance, heat capacity, conductor type, and sag characteristics. For environmental factors, the parameters include: physical influence coefficients of temperature, wind speed, and solar radiation intensity. All physical attribute parameters are derived from equipment nameplate data, factory test reports, and design drawings, and are stored in the attribute fields of the physical attribute layer in the form of structured data. The third layer is the behavior rule layer, which defines the behavioral logic of each entity element under different conditions. For transformers, the behavior rule layer defines the tap changer adjustment rules, such as automatically increasing the tap changer position when the low-voltage side voltage is lower than the set value. For circuit breakers and protection devices, the behavior rule layer defines a state machine model, including: normal monitoring state, protection start state, tripping state, reclosing state, and the transition conditions between each state. For disconnect switches, the behavior rule layer defines the operational interlocking logic between the disconnect switch and the circuit breaker. The definition of the behavior rule layer can be set according to the power industry operating procedures, protection setting sheets, and equipment operation manuals. The fourth layer is the semantic association layer, which defines the semantic relationships between various entity nodes within the substation. The semantic association layer is built based on the substation's main electrical wiring diagram and includes three core relationships: electrical connection relationships, representing the physical electrical connections between devices; protection coordination relationships, representing the logical coordination between protection devices and protected equipment and actuators; and measurement and control association relationships, representing the flow of monitoring signals between measurement and control devices and primary equipment.

[0027] Furthermore, in practical implementation, time registration and spatial coordinate transformation can be performed on real-time multi-source heterogeneous data to align data streams with different sampling frequencies to a unified time reference. The spatial coordinate system of each data point is transformed to the global coordinate system of the engineering multi-dimensional digital twin model. Specifically, in time registration, the BeiDou time service system is used as a unified clock source, and time synchronization devices within the substation provide time synchronization for all sensors and data acquisition equipment. Data collected by each device carries a timestamp. For high-frequency data such as vibration signals and partial discharge signals, a 1-second time window is used to extract characteristic parameters such as peak value and root mean square value as representative values ​​for that moment. For medium-frequency power flow data, the average value within the window is taken. For low-frequency data such as oil chromatography, a hold-before-interpolation strategy is adopted, i.e., the most recent valid value is used until new data arrives. For spatial coordinate transformation, the geodetic coordinate system in the substation design drawings is used as the global coordinate system. For fixed-installation sensors, their installation location coordinates have been pre-measured and entered into the database, and the data is directly associated with the pre-stored coordinates during upload. For mobile acquisition devices, the global coordinates are calculated in real time through their positioning system, and the monitoring data is projected onto the corresponding position in the 3D model. After the transformation is completed, each data point has precise spatial coordinates and can be directly associated with the corresponding device in the 3D model.

[0028] Furthermore, in practical implementation, spatiotemporally aligned multi-source heterogeneous data can be injected into the corresponding layers of the engineering multidimensional digital twin model to drive the pose and appearance state updates of the geometric structure layer, the field quantity distribution updates of the physical attribute layer, and the state transitions and dynamic behavior responses of the behavior rule layer. Specifically, for the geometric structure layer, image data from video surveillance, inspection robots, and drone aerial photography is received. Image recognition algorithms identify the opening and closing positions and abnormal appearance features of the equipment. The recognition results are converted into pose transformation parameters for movable parts in the 3D model, driving the model to move to the corresponding position. Abnormal features are then overlaid and rendered onto the surface of the 3D model as texture maps, achieving real-time updates of the equipment's pose and appearance state. For the physical attribute layer, telemetry data such as voltage, current, power, and frequency from the SCADA system, as well as online monitoring data from oil chromatography online monitoring devices, vibration sensors, gas density sensors, and temperature sensors, are received. These values ​​are written into the attribute fields of the corresponding equipment. For field quantities with spatial distribution characteristics, such as the surface temperature field of equipment collected by an infrared thermal imager, the measured value of each pixel is mapped to the corresponding coordinate position on the surface of a 3D model, forming a continuous spatial distribution cloud map. For environmental data of discrete measurement points, the continuous distribution of the entire field is reconstructed through spatial interpolation algorithms. In the behavior rule layer, action signals from protection devices, equipment position status signals, and control commands are received. The behavior rule layer maintains a finite state machine for each device, defining the various states the device may be in and the transition conditions between states. When external signals or monitoring data meet specific conditions, the behavior rule layer verifies whether the state transition conditions are met. If so, it triggers the device state machine to transition from the current state to the target state, records the state transition time and triggering reason, and outputs the state change event. Finally, in the semantic association layer, the updated entity nodes at each level are associated and bound according to a predefined semantic relationship network, generating a real-time operational status map with causal association weights. This real-time operational status map is the real-time virtual mirror of the power transmission and transformation project. This process can be described in detail by the following steps.

[0029] In this embodiment, a real-time operational status diagram with causal correlation weights is generated, which can be obtained through the following steps: Assign initial causal association weights to the association edges between entity nodes in the semantic relationship network; When the real-time state value of a certain entity node deviates from its steady-state baseline, it is marked as a disturbance source node; Starting from the disturbance source node, the disturbance spreads outward along the association edges of the semantic relationship network, and the disturbance degree value of each downstream association node is determined based on the causal association weight of each association edge. The disturbance values ​​of each node are superimposed along the propagation path to form a causal disturbance distribution map, which in turn generates a real-time operational status map with causal correlation weights.

[0030] In practical implementation, firstly, initial causal association weights are assigned to the edges between entity nodes in the semantic relationship network. That is, the semantic relationship network is constructed based on the substation's main electrical wiring diagram. Entity nodes include: incoming lines, busbars, circuit breakers, disconnectors, transformers, outgoing lines, protection devices, and monitoring and control devices. Edges include: electrical connection edges, protection coordination edges, and monitoring and control correlation edges. For each edge, an initial causal association weight is assigned, with a value between 0 and 1, describing the degree to which changes in the state of upstream nodes affect downstream nodes. The initial causal association weights are preset based on expert experience, taking into account the electrical connection relationships and protection coordination logic of the power transmission and transformation project. Initial causal association weights are stored in the semantic association layer as triples, bound one-to-one with each association edge. Secondly, when the real-time state value of an entity node deviates from its steady-state baseline, it is marked as a disturbance source node. That is, during the continuous operation of the digital twin platform, the real-time state value of each entity node is continuously monitored. For each entity node, a set of steady-state baseline parameters is maintained. These parameters are the state benchmark values ​​and their normal fluctuation range obtained through long-term statistical analysis under normal and stable operating conditions. The steady-state baseline is obtained by extracting the entity node's state time-series data within its historical normal operating window, calculating its mean and standard deviation, using the mean as the steady-state baseline value, and using the mean plus or minus three times the standard deviation as the upper and lower boundaries of the normal fluctuation range. When the deviation of a real-time state value of an entity node from its steady-state baseline exceeds the preset normal fluctuation range, the entity node is determined to have experienced an abnormal disturbance and is marked as a disturbance source node. The marking information includes the globally unique identifier of the disturbance source node, the time of the disturbance, the actual deviation of the real-time state value, and the direction of deviation. The initial disturbance level value is set according to the deviation amplitude; the larger the deviation amplitude, the higher the initial disturbance level value, with a maximum value of 1. Then, starting from the disturbance source node, the disturbance propagation spreads outward along the associated edges of the semantic relationship network. Based on the causal association weights of each associated edge, the disturbance level value of each downstream associated node is determined. That is, a breadth-first propagation strategy is adopted, setting the initial disturbance level value of the disturbance source node to 1, and retrieving all outgoing edges starting from the disturbance source node from the semantic relationship network. For each outgoing edge, the initial causal association weights corresponding to the associated edge and the electrical attenuation factor determined according to the associated edge type are retrieved. The electrical attenuation factor can be preset by expert experience. The disturbance level value of the current node is multiplied by the causal association weights and then by the electrical attenuation factor to obtain the disturbance contribution value of the downstream node. After completing the first hop propagation, if a node receives disturbance contribution values ​​from multiple incoming edges, these contribution values ​​are accumulated and saturated with an upper limit of 1. The accumulated result is used as the disturbance level value of the node.

[0031] Furthermore, in practical implementation, the disturbance levels of each node are superimposed along the propagation path to form a causal disturbance distribution map, which in turn generates a real-time operational status map with causal correlation weights. Specifically, nodes with disturbance levels not lower than a preset ignore threshold are used as new propagation starting points, and the above calculation process is repeated for the next hop. If a visited node is encountered during propagation, its disturbance contribution value is only accumulated without repeated propagation downwards. If a circular path exists in the semantic relationship network, the hop count at the first arrival is used as the reference, and propagation along the loop is not repeated. This hop-by-hop recursive propagation continues until the disturbance levels of all downstream nodes in a given hop are lower than the preset ignore threshold, at which point the propagation process automatically terminates. Finally, each entity node in the semantic relationship network is assigned a disturbance level value, forming a causal disturbance distribution map. This causal disturbance distribution map is then layer-fused and superimposed with a digital twin model that has completed real-time state updates of the geometric structure layer, physical attribute layer, and behavioral rule layer to obtain a real-time operational status map with causal correlation weights. This real-time operational status map serves as a real-time virtual mirror of the power transmission and transformation project.

[0032] In step S3, based on the real-time virtual image, the power transmission and transformation process is simulated and deduced to identify abnormal events and potential weak links in the power transmission and transformation process, and the simulation and deduction results are obtained.

[0033] In this embodiment, based on the real-time virtual image, the power transmission and transformation process is simulated and deduced to identify abnormal events and potential weak points in the power transmission and transformation process, and to obtain the simulation and deduction results. Specifically, this can be achieved through the following steps: Extract a snapshot of the current running state from the real-time virtual image; Using the current running state snapshot as the initial condition, the dual-channel parallel simulation engine is driven to perform the deduction in the behavior rule layer of the multidimensional digital twin model of the project; The dual-channel parallel simulation engine includes: a first channel that runs at a rate proportional to the physical clock to perform device-level state evolution simulations; and a second channel that runs at a rate faster than the physical clock to perform system-level dynamic simulations of power flow and stability. An abnormal event list is extracted from the simulation results of the first channel, and a weak link warning list is extracted from the simulation results of the second channel. The two are then combined as the simulation results.

[0034] In practical implementation, firstly, a snapshot of the current operating state can be extracted from the real-time virtual image. That is, from the real-time virtual image with the causal disturbance distribution map superimposed, the complete operating state information of the power transmission and transformation project at the current moment is extracted to form an operating state snapshot. The operating state snapshot includes: the real-time operating parameters of each power transmission and transformation device; the current operating state of each device; and the current disturbance degree value of each node in the causal disturbance distribution map. The disturbance degree value reflects the degree of causal disturbance impact that each device has experienced at the current moment. Secondly, using the current operating state snapshot as the initial condition, the dual-channel parallel simulation engine is driven to perform inference in the behavior rule layer of the multi-dimensional digital twin model of the project. That is, the extracted current operating state snapshot is loaded into the initial condition of the dual-channel parallel simulation engine, and the two channels start the inference simultaneously based on the same initial state but using different simulation step sizes and simulation models. The simulation process is coordinated by a time synchronization mechanism in the behavior rule layer, ensuring that the two channels run independently at their respective set rates. The dual-channel parallel simulation engine consists of two channels: a first channel running at a rate proportional to the physical clock, performing device-level state evolution simulations; and a second channel running at a super-real-time rate faster than the physical clock, performing system-level dynamic simulations of power flow and stability. Specifically, the first channel runs at a real-time rate, meaning the simulation time is synchronized with physical time, and its device-level state evolution simulation will be detailed in subsequent steps. The second channel, running at a super-real-time rate faster than the physical clock, focuses on system-level power flow distribution and stability dynamic analysis. It uses a simplified power flow calculation model to rapidly simulate the overall operating state of the transmission and transformation system, with the simulation time step typically set to the second level. In the second channel, the power grid topology and power flow data from the operating state snapshot are used as the initial conditions. The simulation simulates the power grid's operating trend over a future period at a super-real-time rate, including the evolution trajectory of voltage at each node, changes in power transmission along each line, and fluctuations in system frequency. The output of the second channel includes: early warning information for system-level weak links, such as: bus voltage over-limit warning, line overload warning, and power angle instability risk warning. Finally, an abnormal event list is extracted from the simulation results of the first channel, and a weak link early warning list is extracted from the simulation results of the second channel. The two are then combined as the simulation result, i.e., the abnormal event list output by the first channel is extracted. The system-level simulation results output by the second channel are analyzed. When the voltage prediction curve of a certain bus is detected to fall below the voltage stability threshold within a preset time window, or the power prediction value of a certain line is detected to exceed the thermal stability limit, a weak link early warning record is generated. The weak link early warning record includes: weak node identifier, early warning type, expected instability time, root cause node information, and suggested focus areas. All weak link early warning records are summarized to form a weak link early warning list. The abnormal event list and the weak link early warning list are merged, and a unified timestamp is added as the simulation result.

[0035] In this embodiment, the device-level state evolution deduction can be implemented using the following steps: For each key power transmission and transformation equipment, extract current characteristic parameters that characterize the health status of the equipment; The current feature parameters are input into a pre-established equipment state prediction model to obtain the equipment state evolution trend over a future period of time; During the evolution simulation, the evolution trend is corrected based on the upstream disturbances reflected in the real-time operational status diagram. When the state evolution trend reaches the alarm threshold within a preset look-ahead time window, an abnormal event record is generated.

[0036] In practical implementation, firstly, for each key power transmission and transformation equipment, current characteristic parameters representing the equipment's health status can be extracted. That is, a set of characteristic parameters effectively representing the current health status of the equipment can be extracted from the real-time virtual image's operational status snapshot. For transformer equipment, the extracted characteristic parameters include: top oil temperature, load current, dissolved gas concentration and growth rate in the oil, and vibration signal amplitude. For circuit breaker equipment, the extracted characteristic parameters include: cumulative number of interruptions, waveform characteristics of the opening and closing coil current, and SF6 gas pressure. For gas-insulated combined electrical equipment, the extracted characteristic parameters include: gas pressure in each chamber, trace moisture content, and partial discharge signal amplitude. These characteristic parameters are combined to form a health status feature vector for the current moment. Secondly, the current characteristic parameters are input into a pre-established equipment status prediction model to obtain the equipment's status evolution trend over a future period. That is, the equipment status prediction model can adopt a neural network model trained based on historical health degradation data. The health status feature vector is used as the model input to run the equipment status prediction model and output the status evolution trend curve of the equipment in the future prospective time window. Then, in the evolution simulation process, the evolution trend is corrected according to the upstream disturbance impact reflected in the real-time operation status map. That is, at each step of the equipment status evolution simulation, the disturbance degree value propagated from the upstream disturbance source node to the equipment through the causal transmission path in the real-time operation status map is obtained. The disturbance level is used as a correction coefficient in the equipment condition prediction model to adjust the slope of the evolution trend curve. The larger the disturbance level, the larger the correction magnitude, and the more the evolution trend leans towards accelerated deterioration. Finally, when the condition evolution trend reaches the alarm threshold within the preset look-ahead time window, an abnormal event record is generated. That is, the corrected equipment condition evolution trend curve is monitored. When the prediction curve shows that the condition value will reach the preset alarm threshold for the first time within the preset look-ahead time window, the alarm threshold can be preset by experimental data analysis. This moment is recorded as the first time the boundary is crossed, and an abnormal event record is generated. The abnormal event record includes the abnormal equipment identifier, abnormal type, predicted boundary crossing time, and suggested preliminary response measures, which are then combined to form an abnormal event list.

[0037] In step S4, based on the preset control rule base and the simulation results, multi-dimensional engineering control instructions are generated and sent to the execution terminal of the power transmission and transformation project to adjust the real-time operating status of the power transmission and transformation project.

[0038] In this embodiment, generating multi-dimensional engineering control instructions based on a preset control rule base and the simulation results can be achieved through the following steps: Extract the abnormal equipment identifier, abnormal type and first boundary crossing time from the abnormal event list from the simulation results, and extract the weak node identifier, instability nucleation time and root cause node information from the weak link early warning list; The extracted information is matched with a preset control rule base at multiple levels. The control rule base includes: equipment control rule cluster, power grid topology adjustment rule cluster, and protection strategy adjustment rule cluster. Based on the rule matching results, a set of coordinated control instructions is constructed; The coordinated control instruction set is converted into engineering control instructions adapted to the execution terminal of physical power transmission and transformation projects.

[0039] In practical implementation, firstly, the abnormal equipment identifier, abnormal type, and first boundary crossing time are extracted from the abnormal event list in the simulation results. Secondly, the weak node identifier, instability nucleation time, and root cause node information are extracted from the weak link early warning list. That is, the abnormal event list and the weak link early warning list are analyzed separately from the simulation results output by the dual-channel parallel simulation. For the abnormal event list, the globally unique identifier, abnormal type code, and predicted first boundary crossing time of each abnormal device are extracted. For the weak link early warning list, the globally unique identifier, predicted instability nucleation time, and root cause node information determined through root cause tracing analysis of each weak node are extracted. Thirdly, the extracted information is matched with a pre-set control rule base at multiple levels. This control rule base includes: equipment control rule clusters, power grid topology adjustment rule clusters, and protection strategy adjustment rule clusters. The control rule base is pre-stored in the digital twin platform and is formulated by experts based on operating procedures and scheduling experience. The equipment control rule cluster handles equipment-level abnormal events, including transformer cooler control rules, on-load tap changer regulation rules, and circuit breaker operation rules. Their priority is determined by the severity of the abnormal event, with rules related to equipment safety protection having the highest priority. The power grid topology adjustment rule cluster handles system-level weak link early warnings, including capacitor bank switching rules, operating mode switching rules, and load transfer rules. Their priority is determined by the degree of impact on power grid stability. The protection strategy adjustment rule cluster adjusts the operating settings and action logic of protection devices, including overload protection setting adjustment rules and distance protection coordination rules. Their priority is determined by the importance of the protection device in the power grid. The rule matching process uses extracted abnormal and weak link information as trigger conditions, matching them step-by-step with rules in the control rule base. Successfully matched rules are sorted by priority and enter the execution queue. Then, based on the rule matching results, a collaborative control instruction set is constructed. Specifically, conflict detection and resolution are performed on successfully matched rules. When multiple rules simultaneously point to the same device but execute in opposite directions, a rule conflict is determined. Conflict detection employs a conflict resolution strategy based on rule priority, retaining higher-priority rules and suspending lower-priority rules. After conflict resolution, the remaining rules are sorted according to time sequence and execution dependencies to form a coordinated control instruction set. This set includes: equipment operating parameter control strategies, power grid topology scheduling strategies, and protection control strategies. Finally, the coordinated control instruction set is converted into engineering management instructions adapted to the physical transmission and transformation engineering execution terminals. Specifically, based on the communication protocols and interface specifications of various execution terminals, the logical instructions in the coordinated control instruction set are converted into issueable engineering management instructions.

[0040] In addition, in specific implementation, control instructions are sent to the execution terminals of the power transmission and transformation project to adjust the real-time operating status of the power transmission and transformation project. That is, the converted project control instructions are sent to the corresponding execution terminals through the industrial communication network of the power transmission and transformation project according to the type and communication protocol of their target execution terminals to adjust the real-time operating status of the power transmission and transformation project.

[0041] like Figure 2 As shown in the diagram, the overall architecture of the digital twin-based power transmission and transformation project management and control system provided in this application includes: a power transmission and transformation project, edge computing nodes, a digital twin platform server, and a power transmission and transformation project execution terminal. Specifically, the system collects operational status data of equipment, power grid topology and power flow data, and engineering environment data from the power transmission and transformation project, and transmits the collected data to the edge computing nodes. The edge computing nodes perform feature extraction and preprocessing on the high-frequency monitoring data, and upload the compressed feature data to the digital twin platform server. The digital twin platform server, based on a pre-built multi-dimensional digital twin model of the project, performs spatiotemporal alignment and semantic fusion of the data to generate a real-time virtual image of the power transmission and transformation project and drive simulation deduction, generating project management and control instructions. The project management and control instructions are issued to the power transmission and transformation project execution terminal, which performs operations such as equipment control, topology adjustment, and protection strategy adjustment, realizing closed-loop management and control of the real-time operational status of the power transmission and transformation project.

[0042] Therefore, this application firstly acquires real-time multi-source heterogeneous data of power transmission and transformation projects by deploying a multimodal sensing network. This reduces the amount of data transmission while preserving key characteristics of equipment status, providing a data foundation for driving the subsequent digital twin model. Secondly, by injecting multi-source heterogeneous data into the digital twin model under a unified spatiotemporal reference and assigning relational edges between entity nodes in the semantic association layer, the generated real-time virtual image can reproduce the explicit operating state of the power transmission and transformation project and reflect the causal transmission path of disturbances in the electrical network, providing a virtual mapping foundation with causal perception capabilities for subsequent simulation and joint control decisions. Thirdly, by using the snapshot of the operating state and the causal disturbance distribution in the real-time virtual image as initial conditions, a dual-channel parallel simulation engine is driven to achieve forward-looking identification of equipment anomalies, providing early warning information with both time margin and root cause orientation for control decisions. Finally, by performing multi-level matching between simulation results and the control rule base, precise closed-loop response to disturbance events and systematic coordinated control of the power transmission and transformation project are achieved.

[0043] In summary, the technical solution adopted in this application can construct a digital twin with causal perception capabilities and drive dual-channel parallel simulation and deduction, thereby improving the accuracy of anomaly prediction and the synergy of multi-dimensional control in power transmission and transformation projects.

[0044] Example 2: This application provides a power transmission and transformation project control system based on digital twins, referencing... Figure 3 As shown in the figure, this is a modular structure diagram of a power transmission and transformation engineering control system based on digital twins according to this embodiment of the present application. The control system includes: The data acquisition module 100 is used to acquire real-time multi-source heterogeneous data of power transmission and transformation projects; The image generation module 200 is used to perform spatiotemporal alignment and semantic fusion on the real-time multi-source heterogeneous data based on the pre-built engineering multi-dimensional digital twin model to generate a real-time virtual image of the power transmission and transformation project. The simulation and deduction module 300 is used to drive the simulation and deduction of the power transmission and transformation process based on the real-time virtual image, identify abnormal events and potential weak links in the power transmission and transformation process, and obtain simulation and deduction results. The instruction generation module 400 is used to generate multi-dimensional engineering control instructions based on a preset control rule base and the simulation results, and to send the engineering control instructions to the execution terminal of the power transmission and transformation project to adjust the real-time operating status of the power transmission and transformation project.

[0045] This application is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0046] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be implemented by a program instructing related hardware. The program can be stored in a computer-readable storage medium, including read-only memory (ROM), random access memory (RAM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), one-time programmable read-only memory (OTPROM), electrically-Erasable Programmable Read-Only Memory (EEPROM), compactdisc read-only memory (CD-ROM) or other optical disc storage, disk storage, magnetic tape storage, or any other computer-readable medium capable of carrying or storing data.

[0047] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus. Unless otherwise specified, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes that element.

Claims

1. A method for the management and control of power transmission and transformation projects based on digital twins, characterized in that, The control method includes the following steps: Acquire real-time multi-source heterogeneous data from power transmission and transformation projects; Based on a pre-built multidimensional digital twin model of the project, the real-time multi-source heterogeneous data is spatiotemporally aligned and semantically fused to generate a real-time virtual image of the power transmission and transformation project. Based on the real-time virtual image, the power transmission and transformation process is simulated and deduced to identify abnormal events and potential weak links in the power transmission and transformation process, and to obtain the simulation and deduction results. Based on the preset control rule base and the simulation results, multi-dimensional engineering control instructions are generated and sent to the execution terminal of the power transmission and transformation project to adjust the real-time operation status of the power transmission and transformation project.

2. The power transmission and transformation engineering management and control method based on digital twin as described in claim 1, characterized in that, The multi-source heterogeneous data includes: operating status data of power transmission and transformation equipment, power grid topology and power flow data, and engineering environment data.

3. The power transmission and transformation engineering management and control method based on digital twin as described in claim 1, characterized in that, Based on a pre-built multi-dimensional digital twin model of the project, spatiotemporal alignment and semantic fusion are performed on the real-time multi-source heterogeneous data to generate a real-time virtual image of the power transmission and transformation project. Specifically, this includes: A pre-constructed multidimensional digital twin model of an engineering project is obtained. The multidimensional digital twin model of an engineering project includes a geometric structure layer, a physical attribute layer, a behavioral rule layer, and a semantic association layer. The layers are associated with each other through a preset mapping relationship. The real-time multi-source heterogeneous data is registered in time and transformed in space to align data streams with different sampling frequencies to a unified time reference, and the spatial coordinate system of each data point is transformed to the global coordinate system of the engineering multidimensional digital twin model. The spatiotemporally aligned multi-source heterogeneous data is injected into the corresponding level of the engineering multidimensional digital twin model to drive the pose and appearance state update of the geometric structure layer, the field quantity distribution update of the physical attribute layer, and the state transition and dynamic behavior response of the behavior rule layer. In the semantic association layer, the updated entity nodes at each level are associated and bound according to the predefined semantic relationship network to generate a real-time operation status diagram with causal association weights. The real-time operation status diagram is a real-time virtual image of the power transmission and transformation project.

4. The power transmission and transformation engineering control method based on digital twin as described in claim 3, characterized in that, Generating a real-time operational status map with causal correlation weights specifically includes: Assign initial causal association weights to the association edges between entity nodes in the semantic relationship network; When the real-time state value of a certain entity node deviates from its steady-state baseline, it is marked as a disturbance source node; Starting from the disturbance source node, the disturbance spreads outward along the association edges of the semantic relationship network, and the disturbance degree value of each downstream association node is determined based on the causal association weight of each association edge. The disturbance values ​​of each node are superimposed along the propagation path to form a causal disturbance distribution map, which in turn generates a real-time operational status map with causal correlation weights.

5. The power transmission and transformation engineering control method based on digital twin as described in claim 4, characterized in that, The initial causal correlation weights are preset by expert experience based on the electrical connection relationships and protection coordination logic of power transmission and transformation projects.

6. The power transmission and transformation engineering management and control method based on digital twin as described in claim 1, characterized in that, Based on the real-time virtual image, the power transmission and transformation process is simulated and deduced to identify abnormal events and potential weaknesses in the power transmission and transformation process. The simulation and deduction results specifically include: Extract a snapshot of the current running state from the real-time virtual image; Using the current running state snapshot as the initial condition, the dual-channel parallel simulation engine is driven to perform the deduction in the behavior rule layer of the multidimensional digital twin model of the project; The dual-channel parallel simulation engine includes: a first channel that runs at a rate proportional to the physical clock to perform device-level state evolution simulations; and a second channel that runs at a rate faster than the physical clock to perform system-level dynamic simulations of power flow and stability. An abnormal event list is extracted from the simulation results of the first channel, and a weak link warning list is extracted from the simulation results of the second channel. The two are then combined as the simulation results.

7. The power transmission and transformation engineering control method based on digital twin as described in claim 6, characterized in that, The device-level refined state evolution simulation specifically includes: For each key power transmission and transformation equipment, extract current characteristic parameters that characterize the health status of the equipment; The current feature parameters are input into a pre-established equipment state prediction model to obtain the equipment state evolution trend over a future period of time; During the evolution simulation, the evolution trend is corrected based on the upstream disturbances reflected in the real-time operational status diagram. When the state evolution trend reaches the alarm threshold within a preset look-ahead time window, an abnormal event record is generated.

8. The power transmission and transformation engineering management and control method based on digital twin as described in claim 1, characterized in that, Based on the preset control rule base and the simulation results, multi-dimensional engineering control instructions are generated, specifically including: Extract the abnormal equipment identifier, abnormal type and first boundary crossing time from the abnormal event list from the simulation results, and extract the weak node identifier, instability nucleation time and root cause node information from the weak link early warning list; The extracted information is matched with a preset control rule base at multiple levels. The control rule base includes: equipment control rule cluster, power grid topology adjustment rule cluster, and protection strategy adjustment rule cluster. Based on the rule matching results, a set of coordinated control instructions is constructed; The coordinated control instruction set is converted into engineering control instructions adapted to the execution terminal of physical power transmission and transformation projects.

9. A power transmission and transformation engineering management and control method based on digital twin as described in claim 8, characterized in that, The coordinated control instruction set includes: equipment operating parameter control strategy, power grid topology scheduling strategy, and protection control strategy.

10. A digital twin-based power transmission and transformation project management and control system, used to execute a digital twin-based power transmission and transformation project management and control method as described in any one of claims 1 to 9, characterized in that, The control system includes: The data acquisition module is used to acquire real-time multi-source heterogeneous data from power transmission and transformation projects. The image generation module is used to perform spatiotemporal alignment and semantic fusion on the real-time multi-source heterogeneous data based on the pre-built multi-dimensional digital twin model of the project, and generate a real-time virtual image of the power transmission and transformation project. The simulation and deduction module is used to drive the simulation and deduction of the power transmission and transformation process based on the real-time virtual image, identify abnormal events and potential weak links in the power transmission and transformation process, and obtain simulation and deduction results; The instruction generation module is used to generate multi-dimensional engineering control instructions based on a preset control rule base and the simulation results, and to send the engineering control instructions to the execution terminal of the power transmission and transformation project to adjust the real-time operating status of the power transmission and transformation project.

Images