Power transmission line loss optimization method and system based on digital twinning

Abstract

The application provides a power transmission line loss optimization method and system based on digital twinning, relates to the technical field of power transmission management, determines a line topology and a state vector through a digital twinning model, constructs an energy consumption interdependence graph of a power transmission line, determines a multi-dimensional feature tensor according to a current coupling relationship between each node in the energy consumption interdependence graph, performs local hotspot evolution on the energy consumption interdependence graph to obtain a loss diffusion subgraph, performs dynamic relaxation checking on a state quantity of the power transmission line to obtain a variable coordination factor, performs optimization feedback on the digital twinning model according to the multi-dimensional feature tensor and the variable coordination factor to obtain a scheduling feedback strategy, determines a multi-level control instruction set according to a scheduling priority of the power transmission line and the scheduling feedback strategy, and performs local loss correction on the power transmission line based on the multi-level control instruction set. The application can realize dynamic evaluation and loss correction of the power transmission line loss based on digital twinning mapping of the power transmission state, so as to improve the operation efficiency of the power transmission line.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of power transmission management, more particularly, the present application relates to a power transmission line loss optimization method and system based on digital twinning. BACKGROUND

[0002] With the continuous expansion of the scale of the power system and the increasing volatility of the load side, the operation efficiency and energy consumption level of the power transmission system, as the core link connecting the power generation end and the load end, have a key impact on the stability and economy of the entire power grid. The power transmission management technology mainly relies on static models and periodic scheduling strategies, which are difficult to accurately perceive the dynamic changes of the line operation state, especially in the face of complex environmental disturbances and random load fluctuations, there are problems such as response lag, coarse optimization granularity, and difficult to effectively inhibit loss.

[0003] Digital twinning technology has become an important technology for realizing the visualization, intelligentization and adaptive control of power transmission lines in power system management. Digital twinning can reflect the behavior characteristics of the physical system under different operating conditions, thereby providing data and model support for loss assessment, dispatch optimization and fault prediction applications. Digital twinning methods attempt to integrate current, voltage, temperature and other sensor data to estimate line energy consumption through a heat consumption calculation model, and assist the decision system in generating operation adjustment strategies. However, existing power transmission line loss optimization methods based on digital twinning still have limited modeling granularity, only modeling at the node level, lack the ability to identify and track high-loss propagation paths in local areas, and the coupling relationship between electrical states is often estimated linearly, lacking dynamic linkage modeling and real-time feedback channels between variables, making it difficult to implement hierarchical scheduling and local fine-tuning of loss according to different line risk levels. Therefore, how to realize dynamic assessment and loss correction of power transmission line loss based on digital twinning mapping of power transmission state to improve the operation efficiency of power transmission lines is a difficult problem faced by the industry. SUMMARY

[0004] The present application provides a power transmission line loss optimization method and system based on digital twinning, which can realize dynamic assessment and loss correction of power transmission line loss based on digital twinning mapping of power transmission state to improve the operation efficiency of power transmission lines.

[0005] In a first aspect, the present application provides a power transmission line loss optimization method and system based on digital twinning, which comprises the following steps:

[0006] Constructing a digital twinning model of the power transmission line, determining the line topology and state vector through the digital twinning model;

[0007] Constructing an energy consumption interdependence graph of the power transmission line through the line topology and the state vector, determining a multi-dimensional feature tensor according to the current coupling relationship between each node in the energy consumption interdependence graph;

[0008] Evolution of the energy consumption interdependence graph is carried out on the local hot spot to obtain a loss diffusion subgraph, and a dynamic relaxation check is carried out on the state quantity of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor;

[0009] The digital twin model is optimized and fed back according to the multi-dimensional feature tensor and the variable coordination factor to obtain a scheduling feedback strategy;

[0010] The scheduling priority of the power transmission line is determined, the multi-level control instruction set is determined according to the scheduling priority and the scheduling feedback strategy, and then the local loss of the power transmission line is corrected based on the multi-level control instruction set.

[0011] In this embodiment, the digital twin model includes a twin perception layer, a coupling modeling layer and a simulation mapping layer, wherein:

[0012] The twin perception layer is used to collect electrical operation parameters, structural parameters and environmental meteorological parameters of the power transmission line to obtain multi-source heterogeneous monitoring data and line topology;

[0013] The coupling modeling layer models the multi-source heterogeneous monitoring data to obtain a state vector of the physical parameters of the power transmission line;

[0014] The simulation mapping layer generates a thermal loss estimation surface based on the state vector, and the thermal loss estimation surface is used to determine the energy consumption interdependence relationship of the power transmission line.

[0015] In this embodiment, the determination of the line topology and the state vector by the digital twin model specifically includes:

[0016] The line topology is determined by the twin perception layer of the digital twin model;

[0017] The state vector is output by the coupling modeling layer of the digital twin model.

[0018] In this embodiment, the construction of the energy consumption interdependence graph of the power transmission line by the line topology and the state vector specifically includes:

[0019] The state vector is mapped to the nodes and edges of the line topology to obtain different attribute tensors;

[0020] The edge weight model is determined according to all the attribute tensors and the energy consumption interdependence relationship of the power transmission line;

[0021] The energy consumption interdependence graph of the power transmission line is constructed based on the Bayesian graph and the edge weight model.

[0022] In this embodiment, the determination of the multi-dimensional feature tensor according to the current coupling relationship between each node in the energy consumption interdependence graph specifically includes:

[0023] extracting an initial state tensor corresponding to each node from the energy-dependent intergraph;

[0024] obtaining current coupling strengths between all nodes in the energy-dependent intergraph, and constructing a current influence matrix according to the current coupling strengths between all nodes;

[0025] supplementing the initial state tensor corresponding to each node through the current influence matrix, and obtaining a multi-dimensional feature tensor.

[0026] In this embodiment, local hotspot evolution is performed on the energy-dependent intergraph to obtain a loss diffusion subgraph, specifically including:

[0027] Performing current density monitoring on the energy-dependent intergraph to obtain a hotspot node set;

[0028] constructing a diffusion domain with a hotspot node in the hotspot node set as the center, and then extracting a thermal inertia coefficient from the diffusion domain;

[0029] determining a current migration trend based on the thermal inertia coefficient;

[0030] diffusion evolution is performed on the current migration trend to obtain a loss diffusion subgraph.

[0031] In this embodiment, based on the loss diffusion subgraph, dynamic relaxation verification is performed on the state quantity of the power transmission line to obtain a variable coordination factor, specifically including:

[0032] constructing a time sequence state matrix based on the loss diffusion subgraph;

[0033] performing relaxation boundary calculation on the time sequence state matrix to obtain a tension boundary and a relaxation boundary of the state quantity change of the power transmission line;

[0034] using difference evolution to correlate the tension boundary and the relaxation boundary to obtain a coordinated response fluctuation of the state quantity change of the power transmission line;

[0035] extracting a variable coordination factor from the coordinated response fluctuation.

[0036] In this embodiment, determining the scheduling priority of the power transmission line specifically includes:

[0037] obtaining historical fault records and real-time state quantities of the power transmission line;

[0038] constructing a multi-factor risk assessment model based on the historical fault records;

[0039] performing risk assessment on the real-time state quantities through the multi-factor risk assessment model to obtain the scheduling priority of the power transmission line.

[0040] In the embodiment, the state quantity is a set of physical quantities representing the real-time operation state of the power transmission line, and the state quantity includes electrical parameters and thermal parameters.

[0041] In a second aspect, the application provides a power transmission line loss optimization system based on digital twinning, which is used to execute a power transmission line loss optimization method based on digital twinning.

[0042] A line state twinning module is configured to construct a digital twinning model of the power transmission line, and determine a line topology and a state vector through the digital twinning model.

[0043] A feature tensor extraction module is configured to construct an energy consumption interdependence graph of the power transmission line through the line topology and the state vector, and determine a multi-dimensional feature tensor according to the current coupling relationship between each node in the energy consumption interdependence graph.

[0044] A hotspot evolution verification module is configured to perform local hotspot evolution on the energy consumption interdependence graph to obtain a loss diffusion subgraph, and perform dynamic relaxation verification on the state quantity of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor.

[0045] A strategy optimization feedback module is configured to perform optimization feedback on the digital twinning model according to the multi-dimensional feature tensor and the variable coordination factor to obtain a scheduling feedback strategy.

[0046] A loss correction execution module is configured to determine a scheduling priority of the power transmission line, determine a multi-level control instruction set according to the scheduling priority and the scheduling feedback strategy, and then perform local loss correction on the power transmission line based on the multi-level control instruction set.

[0047] The technical scheme provided by the embodiments of the application has the following beneficial effects:

[0048] The digital twinning model of the power transmission line is constructed, the line topology and the state vector are determined through the digital twinning model, the energy consumption interdependence graph of the power transmission line is constructed through the line topology and the state vector, the multi-dimensional feature tensor is determined according to the current coupling relationship between each node in the energy consumption interdependence graph, the local hotspot evolution is performed on the energy consumption interdependence graph to obtain the loss diffusion subgraph, the dynamic relaxation verification is performed on the state quantity of the power transmission line based on the loss diffusion subgraph to obtain the variable coordination factor, the optimization feedback is performed on the digital twinning model according to the multi-dimensional feature tensor and the variable coordination factor to obtain the scheduling feedback strategy, the scheduling priority of the power transmission line is determined, the multi-level control instruction set is determined according to the scheduling priority and the scheduling feedback strategy, and then the local loss correction is performed on the power transmission line based on the multi-level control instruction set.

[0049] It can be seen that in the present application, the dynamic evaluation and loss correction of the power transmission line loss can be realized based on the digital twin mapping of the power transmission state. First, by constructing a digital twin-like model of the power transmission line, the line topology structure and state vector are extracted, so that the system can synchronize the running state of the physical power transmission line in the virtual space in real time, and a unified framework connecting perception, simulation and scheduling is constructed, which is conducive to the subsequent fusion expression of multi-source data in complex power transmission scenarios. And by constructing the energy dependence graph based on the line topology and state vector, and combining the current coupling relationship between the nodes in the graph to determine the multi-dimensional feature tensor, the distributed current flow and local loss response relationship can be encoded into a structured data model, realizing the graph modeling of the energy transmission path and loss behavior in the power transmission system, which is conducive to accurately identifying high-loss paths and abnormal load areas. Second, by extracting the loss diffusion subgraph and performing dynamic relaxation verification on the state quantity, the variable coordination factor is extracted, which can timely identify and quantify the propagation trend and linkage strength of local electrical disturbance in the system, improving the spatiotemporal sensitive modeling capability of the line operation characteristics. Then, the digital twin model is fed back and optimized by using the multi-dimensional feature tensor and the variable coordination factor, and a scheduling feedback strategy is generated, so that the digital twin model has the active scheduling core of adaptive regulation and control function, which is conducive to realizing the rapid perception and intelligent response capability of abnormal load and high-loss trend. Finally, combined with the scheduling feedback strategy and the line scheduling priority, a multi-level control instruction set is generated and applied to the power transmission system for local loss correction, which can effectively reduce unnecessary power loss and improve the rationality of line power distribution and the response speed of scheduling execution.

[0050] In summary, the technical scheme adopted in the present application can realize dynamic evaluation and loss correction of power transmission line loss based on digital twin mapping of power transmission state, so as to improve the operation efficiency of the power transmission line. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the prior art, the drawings needed to be used in the embodiments or prior art description will be briefly introduced. Obviously, the drawings in the following description are only the embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0052] Figure 1 is a flowchart of the power transmission line loss optimization method based on digital twin provided by the present application;

[0053] Figure 2 is an exemplary flowchart for determining the energy dependence graph according to the present application;

[0054] Figure 3 is an exemplary flowchart for determining the loss diffusion subgraph according to the present application;

[0055] Figure 4 is a module structure diagram of an optimization system provided according to the present application. DETAILED DESCRIPTION

[0056] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative work fall within the scope of protection of the present application.

[0057] The embodiments of the present application provide a power transmission line loss optimization method and system based on digital twinning, the core of which is to construct a digital twinning model of a power transmission line, to determine a line topology and a state vector through the digital twinning model; to construct an energy consumption interdependence graph of the power transmission line through the line topology and the state vector, to determine a multi-dimensional feature tensor according to a current coupling relationship between each node in the energy consumption interdependence graph; to perform local hotspot evolution on the energy consumption interdependence graph to obtain a loss diffusion subgraph, to perform dynamic relaxation checking on state quantities of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor; to perform optimization feedback on the digital twinning model according to the multi-dimensional feature tensor and the variable coordination factor to obtain a scheduling feedback strategy; to determine a scheduling priority of the power transmission line, to determine a multi-level control instruction set according to the scheduling priority and the scheduling feedback strategy, and to perform local loss correction on the power transmission line based on the multi-level control instruction set.

[0058] Embodiment one, in order to better understand the above technical solutions, the above technical solutions will be described in detail below in combination with the drawings in the specification and specific embodiments, referring to Figure 1 The figure is an exemplary flowchart of a power transmission line loss optimization method based on digital twinning according to the embodiments of the present application, the optimization method comprising the following steps:

[0059] In step S1, a digital twinning model of a power transmission line is constructed, and a line topology and a state vector are determined through the digital twinning model.

[0060] In the present embodiment, the digital twinning model comprises a twinning perception layer, a coupling modeling layer and a simulation mapping layer, wherein:

[0061] The twinning perception layer is used to collect electrical operation parameters, structural parameters and environmental meteorological parameters of the power transmission line to obtain multi-source heterogeneous monitoring data and a line topology;

[0062] The coupling modeling layer models the multi-source heterogeneous monitoring data to obtain a state vector of physical parameters of the power transmission line;

[0063] The simulation mapping layer generates a heat loss estimation surface based on the state vector, and the heat loss estimation surface is used to determine the energy consumption dependence relationship of the power transmission line.

[0064] In a specific implementation, first, in the twin perception layer, current sensors, voltage sensors, temperature and humidity sensors, and wind speed and direction instruments are deployed at key nodes (such as the top of the tower and the connection point of the conductor) of the power transmission line to collect electrical operating parameters, structural parameters, and environmental meteorological parameters in real time, and the line topology is extracted in combination with the design drawings of the power transmission line, and the line topology is stored in the twin perception layer through GIS, wherein the electrical operating parameters include the spatial position, electrical load, and meteorological factors of the power transmission line, and the meteorological factors refer to environmental parameters outside the power transmission line, including temperature and humidity, wind speed, and wind direction; then, in the coupled modeling layer, the multi-source heterogeneous monitoring data are aligned by time, and input vectors are constructed through data cleaning and normalization operations, and then the input vectors are input into a regression neural network, and a state vector is obtained through the output of the regression neural network, wherein the state vector is an index describing the conductor temperature rise, current density, and voltage drop; finally, in the simulation mapping layer, a heat loss calculation model based on Bayesian regression is used to calculate the heat loss of the state vector, and a heat loss estimation surface is obtained, wherein the heat loss estimation surface is a visual energy consumption distribution model formed by taking the spatial position, electrical load, and meteorological factors as independent variables and the heat loss power per unit length of the power transmission line as the dependent variable, and the heat loss estimation surface is used to determine the energy consumption dependence relationship of the power transmission line.

[0065] It should be noted that the digital twin model in the present application is a visual model of a multi-layer heterogeneous fusion structure, has integrated perception, modeling, and prediction capabilities, and can form a holographic mapping of the state of the power transmission line; the digital twin model includes a twin perception layer, a coupled modeling layer, and a simulation mapping layer, wherein: the twin perception layer is used for high-frequency collection and unified expression of multi-source data, to ensure the spatiotemporal consistency of the input information; the coupled modeling layer can realize functional mapping between the perception data and the physical state, and is the core that connects the real line and the simulation copy; the simulation mapping layer introduces a loss estimation surface based on Bayesian regression, which is beneficial to improving the accuracy of heat energy loss estimation.

[0066] In the present embodiment, the line topology and the state vector can be determined by the digital twin model in the following manner, that is:

[0067] The line topology is determined by the twin perception layer of the digital twin model.

[0068] The state vector is output by the coupled modeling layer of the digital twin model.

[0069] In a specific implementation, first, in the twin perception layer of the digital twin model, a line topology is extracted by using GIS, the line topology is graph structure data formed by taking structural nodes of a power transmission line as nodes and taking connection relationships of the power transmission line as edges, and is used to describe spatial structures and connection relationships between nodes (such as towers and connection points) in the power transmission line; second, in the coupled modeling layer of the digital twin model, a state vector is obtained by using a regression neural network, and is used to describe electrical states of each node, such as conductor temperature rise, current density, and voltage drop.

[0070] In step S2, an energy dependence graph of the power transmission line is constructed by using the line topology and the state vector, and a multi-dimensional feature tensor is determined according to current coupling relationships between nodes in the energy dependence graph.

[0071] Preferably, in the embodiment, reference is made to FIG. 2, which is an exemplary flowchart for determining an energy dependence graph according to the present application, and the energy dependence graph of the power transmission line can be constructed by using the line topology and the state vector in the embodiment by using the following steps: Figure 2

[0072] In step S21, the state vector is mapped to nodes and edges of the line topology to obtain different attribute tensors;

[0073] In step S22, an edge weight model is determined according to all attribute tensors and energy dependence relationships of the power transmission line;

[0074] In step S23, the energy dependence graph of the power transmission line is constructed based on a Bayesian graph and the edge weight model.

[0075] ​In a specific implementation, firstly, the voltage sag in the state vector can be mapped to the nodes of the line topology by a spatiotemporal data fusion algorithm, the current density and the conductor temperature rise can be mapped to the edges of the line topology, and an attribute tensor can be extracted from the nodes and edges of the mapped topology by an attribute extraction function. The attribute tensor refers to a set of multi-dimensional vector data for representing the physical state of the nodes or edges, including a node voltage tensor, an edge current density tensor, and an edge temperature rise tensor, for describing the electrical state characteristics and their change trends at different topology nodes and edges. Preferably, the attribute extraction function can use a multi-dimensional feature fusion function, which is beneficial to accurate extraction of state characteristics under different time scales and spatial distributions and improves the representation ability of the node and edge features in the subsequent graph model. Then, the edge energy consumption can be obtained by summing the current density tensor and the conductor temperature rise tensor in the attribute tensor and then dividing by the edge length. The edge energy consumption refers to the energy loss level of the unit length of the power transmission line under different operating states. The product of the edge energy consumption and the voltage tensor of the node is calculated, and the products calculated for all nodes are constructed as an edge weight model. Preferably, the edge weight model can be constructed by using a parameter fitting function based on multivariate nonlinear regression, which is beneficial to comprehensively considering the influence of multi-source state data on energy consumption changes and realizing the adaptability of edge weight calculation. Finally, the electrical state (such as the voltage tensor) of each node can be regarded as an observed variable in a conditional probability distribution, and the observed variable can be iteratively updated based on the edge weight model by using a Bayesian information criterion (BIC) optimization algorithm, so as to obtain a stable Bayesian energy consumption graph, which can be used as the energy consumption interdependence graph of the power transmission line.

[0076] It should be noted that the energy consumption interdependence graph in the embodiment is a graph model that integrates topology structure, electrical parameters, and energy consumption characteristics, and is used to describe the energy transmission path and coupling strength between nodes in the power transmission line under a specific operating state. In the embodiment, the Bayesian modeling mechanism is introduced, so that each edge weight in the graph has statistical interpretability and evolution update capability.

[0077] In the embodiment, the multi-dimensional feature tensor can be determined according to the current coupling relationship between the nodes in the energy consumption interdependence graph in the following manner:

[0078] An initial state tensor corresponding to each node is extracted from the energy consumption interdependence graph.

[0079] The current coupling strength between all nodes in the energy consumption interdependence graph is obtained, and a current influence matrix is constructed according to the current coupling strength between all nodes.

[0080] The initial state tensor corresponding to each node is supplemented in multiple channels by using the current influence matrix, and a multi-dimensional feature tensor is obtained.

[0081] In specific implementation, firstly, using nodes in the energy consumption inter-graph as index units, the probability distribution of observed variables in each index unit is extracted using a state inversion function, and this probability distribution is used as the initial state tensor, thus obtaining the initial state tensor corresponding to each node. The initial state tensor refers to a multi-dimensional vector data set reflecting the node's voltage level, current state, and historical dependency weights under current operating conditions. Preferably, the state inversion function can employ an approximate inference algorithm, which is beneficial for accurately recovering the node's initial state under incomplete observation conditions. Then, by calculating the normalized ratio between the current density tensor corresponding to each edge in the energy consumption inter-graph and its associated edge energy consumption value, the current coupling strength between adjacent nodes is obtained. The current coupling strength measures the impact of a node's current change on the energy consumption of its adjacent nodes. The indicators of energy consumption change are analyzed, and a current influence matrix is ​​constructed using all nodes as the index unit. Each element in the current influence matrix represents the current coupling strength. Finally, the current coupling strength of each node in the current influence matrix is ​​used as a weighting coefficient to perform weighted aggregation on the initial state tensors of all nodes adjacent to the target node. This aggregation result is then multi-channel concatenated with the initial state tensor of the target node itself to obtain a multi-dimensional feature tensor containing local topological association features and node state features. The multi-dimensional feature tensor is a set of feature vectors reflecting information channels such as voltage, current, and temperature rise, which is used to enhance the model's ability to perceive the context of node states and characterize energy consumption evolution trends. Preferably, the weighted aggregation can be implemented using graph convolution operations, which is beneficial for improving the directionality and discriminability of feature supplementation.

[0082] It should be noted that the multidimensional feature tensor in this application refers to a high-dimensional parameter set that integrates node state information and factors influencing coupled nodes. It can describe an extended expression of the electrical state of a single node. The current coupling relationship refers to the extent to which a change in the current of one node will cause a change in the current, voltage, or other operating parameters of other nodes. It is used to characterize the degree of electrical interdependence between nodes. By constructing a multidimensional feature tensor, it is beneficial to enhance the state representation capability of the digital twin system.

[0083] In step S3, the energy consumption interdependence graph is subjected to local hotspot evolution to obtain a loss diffusion subgraph. Based on the loss diffusion subgraph, the state variables of the transmission line are dynamically relaxed to obtain the variable coordination factor.

[0084] Preferably, in this embodiment, reference Figure 3 As shown, this figure is an exemplary flowchart for determining the loss diffusion subgraph according to the present application. In this embodiment, the loss diffusion subgraph is obtained by performing local hotspot evolution on the energy consumption inter-graph. The specific steps are as follows:

[0085] In step S31, the current density of the energy consumption interdependence graph is monitored to obtain a set of hotspot nodes;

[0086] In step S32, a diffusion domain is constructed with the hotspot nodes in the set of hotspot nodes as the center, and a thermal inertia coefficient is extracted from the diffusion domain;

[0087] In step S33, a current migration trend is determined based on the thermal inertia coefficient;

[0088] In step S34, the current migration trend is subjected to diffusion evolution to obtain a loss diffusion subgraph.

[0089] In a specific implementation, first, a mean joint evaluation model of a sliding time window algorithm is combined with the current coupling between nodes in the energy consumption interdependence graph to set a current density threshold, and the sliding time window algorithm is used to calculate the current density response strength of each node in the energy consumption interdependence graph in a unit time window, and the nodes corresponding to the current density response strength greater than the current density threshold in the unit time window are all regarded as hotspot nodes, and a set composed of all the hotspot nodes is regarded as a set of hotspot nodes. Second, the coupling nodes of the hotspot nodes are determined through the current coupling relationship, and a diffusion domain containing the hotspot nodes and the coupling nodes is constructed with the hotspot nodes in the set of hotspot nodes as the center. In the diffusion domain, the thermal inertia coefficient is obtained based on the fitting of the heat diffusion equation. The thermal inertia coefficient is a parameter representing the response lag degree of the local region under the input of continuous current, and is used to quantify the heat diffusion capacity and energy consumption aggregation trend of the hotspot nodes. Then, the thermal inertia coefficient in the diffusion domain is jointly modeled with the current density tensor, and the current migration probability function is constructed in combination with the current coupling strength between nodes. If there are adjacent nodes with relatively low thermal inertia coefficient and high current coupling strength in a diffusion domain, it is determined that it is the potential migration direction of the current, and then the current migration trend is formed. The current migration trend is a graph model describing the dynamic migration path of the current in the topology structure. Preferably, the process of determining the current migration trend can be realized through a time series graph neural network, which is beneficial to dynamically reflecting the current redistribution behavior under a complex topology. Finally, the diffusion iteration function can be used to simulate the migration process of the current between adjacent nodes in a unit time step based on the current migration trend, and the local energy consumption value of the node is updated in each iteration, and the loss threshold is set according to the actual material and service life of the power transmission line. All local energy consumption values higher than the loss threshold in three continuous diffusion rounds are regarded as loss nodes, and then a graph composed of all the loss nodes is regarded as a loss diffusion subgraph.

[0090] It should be noted that the local hotspot evolution in the embodiment refers to constructing a local area model highly representative of actual power loss by identifying the hotspot area in the power transmission line map and simulating the heat diffusion trend; the thermal inertia coefficient is used to measure the delay degree of energy transfer and temperature rise change of the node per unit time, reflects the thermal response ability of the node under different operating states, and the node with high thermal inertia usually has high environmental stability, but is more likely to cause energy accumulation under overload conditions; in addition, the loss diffusion subgraph in the application is a graph model reflecting the thermal coupling relationship between different nodes in the energy transfer process.

[0091] In the embodiment, the state quantity is a set of physical quantities representing the real-time operating state of the power transmission line, and the state quantity includes electrical parameters and thermal parameters.

[0092] It should be noted that the state quantity includes electrical parameters and thermal parameters; wherein the electrical parameters include voltage, current, active power, reactive power, power factor, conductance, reactance, resistance, etc., for reflecting the real-time power transmission efficiency and load distribution of the power transmission line in the power grid; the thermal parameters include line conductor surface temperature, ambient temperature, thermal inertia coefficient, thermal conductivity, radiation heat dissipation rate, etc., for characterizing the heat loss and temperature rise effect caused by current passing through the line during operation; the above state quantities can be obtained by real-time collection of sensors and synchronous updating by digital twin technology.

[0093] In the embodiment, the state quantity of the power transmission line is dynamically checked based on the loss diffusion subgraph to obtain the variable coordination factor, which can be obtained in the following manner, that is:

[0094] A time sequence state matrix is constructed based on the loss diffusion subgraph;

[0095] The time sequence state matrix is subjected to relaxation boundary calculation to obtain the tension boundary and relaxation boundary of the state quantity change of the power transmission line;

[0096] The tension boundary and the relaxation boundary are subjected to trend correlation by difference evolution to obtain the coordinated response fluctuation of the state quantity change of the power transmission line;

[0097] The variable coordination factor is extracted from the coordinated response fluctuation.

[0098] In a specific implementation, first, state variables of the loss diffusion subgraph can be extracted through a multivariate time series reconstruction technology, and then the extracted state variables, i.e., current density, conductor temperature rise and voltage drop, are constructed into a time series state matrix through a state aggregation encoding method, wherein the time series state matrix represents the dynamic change characteristics of the state variables of each key node in the diffusion region over time; second, the time series state matrix can be calculated by interval estimation based on change point detection to obtain the tension boundary and relaxation boundary in the state variable change process of the power transmission line, wherein the tension boundary refers to the maximum response upper limit when the node state variable appears a continuous rising trend in a continuous time interval, which is used to represent the fluctuation boundary in the state tension accumulation process, and the relaxation boundary refers to the minimum response lower limit when the node state variable enters a falling trend after the peak value, which is used to depict the response falling range in the state release stage, preferably, the interval estimation based on change point detection can use a joint estimation method based on the range statistics, local slope change rate and change direction in the sliding interval, which is beneficial to accurately identify the tension and relaxation turning points in the state variable change process; then, the trend correlation of the tension boundary and the relaxation boundary can be performed by using a difference evolution algorithm to construct a difference time series of the tension boundary and the relaxation boundary, extract the direction of the state variable change in the continuous time window, and then perform population initialization, mutation, crossover and selection operations on the nonlinear change trend between the tension boundary and the relaxation boundary sequence according to the direction through the difference evolution algorithm to obtain the fluctuation curve of the transition of the tension boundary to the relaxation boundary, and then the fluctuation curve is used as the coordinated response fluctuation of the state variable change of the power transmission line, preferably, the current influence between nodes can be introduced in the above difference evolution process, which is beneficial to improve the spatial consistency of the trend fitting; finally, the response amplitude, phase synchronicity and change direction of the state variable in the tension and relaxation boundary alternation process are extracted from the coordinated response fluctuation, and the response amplitude, phase synchronicity and change direction correlation are aggregated through principal component analysis, and then the aggregation result is used as a variable coordination factor, wherein the variable coordination factor is a quantitative parameter representing the change intensity of the state variable of the power transmission line in the dynamic fluctuation process, preferably, a time series weighting function can be introduced in the process of aggregating the variable coordination factor to enhance the sensitivity to the state variable mutation.

[0099] It should be noted that the variable coordination factor in the present application is a quantitative parameter representing the change intensity of the state quantity of the power transmission line in the dynamic fluctuation process. The variable coordination factor describes the response synchronism, trend consistency and amplitude correlation of the state quantity in the relaxation boundary fluctuation. The internal response law of multiple key variables in the operation state of the power transmission line can be revealed through the variable coordination factor, which assists in identifying abnormal linkage behaviors in local areas. The variable coordination factor can be used to identify high correlation state variable combinations, and provide multi-dimensional joint control basis for the dispatching system. In addition, the dynamic relaxation verification in the present embodiment refers to a process for identifying abnormal amplification or buffering characteristics of the state quantity of the power transmission line under the influence of disturbance. The evolution trajectory between the tension boundary and the relaxation boundary of the state data in the historical time sequence can be analyzed to effectively capture the dynamic overload or inhibition characteristics of the line state. The coordinated response fluctuation refers to the fluctuation curves of multiple state variables under the same disturbance event, which can represent current rise accompanied by temperature rise, voltage fluctuation accompanied by frequency jump, etc.

[0100] In step S4, the digital twin model is optimized and fed back according to the multi-dimensional feature tensor and the variable coordination factor, to obtain a dispatch feedback strategy.

[0101] In the present embodiment, the digital twin model is optimized and fed back according to the multi-dimensional feature tensor and the variable coordination factor, to obtain a dispatch feedback strategy. Specifically, the following methods can be used:

[0102] The multi-dimensional feature tensor and the variable coordination factor are matched and analyzed to obtain a feedback input set;

[0103] The feedback input set is subjected to heat loss response simulation by the simulation mapping layer of the digital twin model, to obtain a thermoelectric coupling response value;

[0104] The operation constraint condition of the power transmission line is subjected to dispatch feedback according to the thermoelectric coupling response value, to generate a dispatch feedback strategy.

[0105] In a specific implementation, first, the voltage, current density, temperature rise and other multi-source state characteristics and variable coordination factors in the multi-dimensional feature tensor are linearly combined through a feature weighting fusion algorithm, and then the combined vector set is taken as a feedback input set, wherein the feedback input set represents the comprehensive response characteristics of the multi-state variables in the current power transmission line under the synergistic evolution; then, the digital twin model performs multi-variable thermal-electric coupling modeling on the feedback input set through a simulation mapping layer, solves the thermal conduction equation and the current distribution model based on finite element analysis, and simulates the thermal loss evolution process under different state variable combinations, wherein in the simulation process, the multi-dimensional state variables in the feedback input set are input as boundary conditions to the simulation mapping layer, the temperature rise gradient, current density distribution and energy consumption accumulation per unit time corresponding to each topology node and edge are calculated, and finally the thermal-electric coupling response value is obtained, which represents the thermal loss response strength of the power transmission line; finally, the operation constraint conditions of the power transmission line are dynamically associated and analyzed according to the thermal-electric coupling response value, and the risk sections of thermal abnormalities, uneven power flow and sudden increase of energy consumption are identified, and a scheduling feedback strategy is generated through the risk labels of the risk sections, that is, the strategy set includes parameter adjustment instructions, local load transfer and energy distribution priority, which can be used for local energy consumption suppression to guide the operation adjustment of the power transmission system.

[0106] It should be noted that in the present application, the generation of the scheduling feedback strategy is realized by combining the current operation state of the power transmission line with the dynamic association characteristics of the state variables, which can realize early intervention before the state abnormality is fully evolved. In addition, the finite element analysis introduced in the simulation mapping layer enables the simulation process to have spatial resolution and boundary coupling characteristics, which can effectively depict the micro dynamic characteristics of energy consumption diffusion and temperature rise evolution under complex topological structure, and provide fine basic data support for parameter configuration of the scheduling feedback strategy, thereby facilitating to improve the predictability and controllability of the digital twin model in intelligent scheduling of the power system.

[0107] In step S5, the scheduling priority of the power transmission line is determined, the multi-level control instruction set is determined according to the scheduling priority and the scheduling feedback strategy, and then the local loss correction of the power transmission line is performed based on the multi-level control instruction set.

[0108] In the present embodiment, the scheduling priority of the power transmission line can be determined in the following manner, that is:

[0109] Obtain the historical fault records and real-time state variables of the power transmission line;

[0110] Construct a multi-factor risk assessment model based on the historical fault records;

[0111] Perform risk assessment on the real-time state variables through the multi-factor risk assessment model to obtain the scheduling priority of the power transmission line.

[0112] In particular implementation, first, the historical fault records are obtained from the dispatching database of the power transmission line, wherein the historical fault records include the number of tripping, overload alarm, cable heating warning, joint aging record, etc., the historical fault records of the past 3 years can be obtained, and the state quantity (electrical parameter and thermal parameter) of the power transmission line is obtained in real time through the digital twin model to obtain the real-time state quantity; then, a multi-factor risk assessment model is constructed based on the historical fault records such as fault frequency, influence range, recovery time, and equipment aging degree; preferably, the model can be established based on the analytic hierarchy process, so that each fault type has a unified risk scoring rule; finally, the real-time state quantity is taken as input data and substituted into the multi-factor risk assessment model to perform dynamic risk scoring to obtain the risk level label (such as high, medium, and low) of the power transmission line, and the dispatching priority of the power transmission line is divided according to the risk level label, that is, the higher the risk level label, the earlier the priority, and the lower the risk level label, the later the priority.

[0113] It should be noted that the multi-factor risk assessment model in the embodiment is the core of the dispatching priority calculation, and the historical fault records and the real-time state quantity can be fused and processed, which is beneficial to avoid the problem of relying only on real-time parameters and ignoring historical hidden dangers.

[0114] In the embodiment, the multi-level control instruction set can be determined according to the dispatching priority and the dispatching feedback strategy in the following manner, that is:

[0115] The control level is divided according to the dispatching priority, and then the response parameter set is determined according to the corresponding control level;

[0116] The control instruction is determined through the corresponding response parameter set, and then the multi-level control instruction set is obtained.

[0117] In specific implementation, first, high-risk, medium-risk and low-risk three levels are divided according to scheduling priority, and the response parameter set of the corresponding control level is set according to the actual power transmission demand and operation experience of the power transmission line, the response parameter set including adjustment amplitude, control response time, control node range, etc., for example: high-risk requires large adjustment amplitude, short response time, and wider control coverage, and low-risk can use slight adjustment or observation; then, the response parameter set corresponding to each control level is mapped by the parameter adjustment instruction in the scheduling feedback strategy, local load transfer and energy distribution priority, in actual implementation, the scheduling feedback strategy and the response parameter set can be matched by a control strategy generation algorithm based on rule matching and optimization search to generate corresponding control instructions, and the control instructions generated by all risk levels are obtained by the above method, and then the set of control instructions generated by all risk levels is used as a multi-level control instruction set to guide the accurate scheduling of the power transmission line under different risk levels; it should be noted that the multi-level control instruction set can realize hierarchical response and dynamic adjustment of the operation state of the power transmission system, improve the scheduling flexibility and risk prevention and control capability of the system through differentiated control strategies, and is conducive to realizing the safe and stable operation of the power transmission system under various complex operation scenarios such as high-voltage load and uncertain weather.

[0118] It should be noted that in the present application, the local loss correction of the power transmission line based on the multi-level control instruction set means that the control instructions generated for different risk levels in the multi-level control instruction set are used to differentially adjust the local area with thermal abnormalities, current surges or energy deviation in the power transmission line, including: adjusting the current distribution ratio of the related node, changing the load path of the node, optimizing the power flow configuration in the scheduling period, etc.; wherein the local loss correction can not only alleviate the energy aggregation problem of the hot spot node, but also reduce the overall operation temperature rise level of the system.

[0119] In specific implementation, first, the key nodes and edges of the power transmission line that need to be corrected are selected according to the corresponding control level in the multi-level control instruction set; then, according to the response parameters such as adjustment amplitude, control response time and control node range contained in the control instruction, the corresponding local regulation operation is performed, for example: rapid load reduction is applied to the high-risk area, the current distribution ratio between conductors is adjusted, or the standby channel is enabled for energy diversion; for medium and low risk areas, methods such as delayed response, slight adjustment or observation waiting can be used to control the energy evolution trend; finally, the thermal loss change of each node after regulation is monitored in real time, and the correction result is fed back to the digital twin model for dynamic updating, realizing the suppression of local abnormal energy consumption behavior and the optimization of power transmission system energy efficiency.

[0120] In addition, it should be noted that in the present application, through the fusion analysis of the multi-dimensional feature tensor and the variable coordination factor, not only the fine identification ability of the state change of the power transmission line is improved, but also the modeling and prediction of the energy consumption behavior under complex operating state are realized through the construction of the thermal-electric coupling response mechanism, so that the dispatching feedback strategy can have stronger real-time performance.

[0121] In summary, the technical scheme adopted in the present application can realize dynamic evaluation and loss correction of power transmission line loss based on digital twin mapping of power transmission state, so as to improve the operating efficiency of the power transmission line.

[0122] In embodiment two, the present application provides a power transmission line loss optimization system based on digital twin, referring to Figure 4 the figure is a module structure diagram of the optimization system provided by the present application, and the optimization system comprises:

[0123] a line state twin module 100 for constructing a digital twin model of the power transmission line, determining the line topology and the state vector through the digital twin model;

[0124] a feature tensor extraction module 200 for constructing an energy consumption interdependence graph of the power transmission line through the line topology and the state vector, determining a multi-dimensional feature tensor according to the current coupling relationship between each node in the energy consumption interdependence graph;

[0125] a hotspot evolution verification module 300 for locally evolving the energy consumption interdependence graph to obtain a loss diffusion subgraph, dynamically relaxing and verifying the state quantity of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor;

[0126] a strategy optimization feedback module 400 for optimizing and feeding back the digital twin model according to the multi-dimensional feature tensor and the variable coordination factor to obtain a dispatching feedback strategy;

[0127] a loss correction execution module 500 for determining the dispatching priority of the power transmission line, determining a multi-level control instruction set according to the dispatching priority and the dispatching feedback strategy, and then locally correcting the loss of the power transmission line based on the multi-level control instruction set.

[0128] The computer program instructions can also be loaded onto a computer, other programmable data processing apparatus, or other processing devices to cause a series of operational steps to be performed on the computer, other programmable apparatus or other devices to produce a computer implemented process such that the instructions which execute on the computer or other programmable apparatus provide processes for implementing the functions specified in the flowchart block or blocks. Figure 1 The flowchart and / or block diagram in the working examples illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments. In this regard, each flowchart block and / or block in the flowcharts and / or block diagrams can represent a module, segment, or portion of code, which comprises one or more executable Figure 1 The flowchart and / or block diagram in the working examples illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments. In this regard, each flowchart block and / or block in the flowcharts and / or block diagrams can represent a module, segment, or portion of code, which comprises one or more executable

[0129] Those skilled in the art can understand that all or part of the steps in the above-mentioned embodiments can be completed by instructing relevant hardware through a program, and 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), Compact Disc Read-Only Memory (CD-ROM) or other optical disk storage, magnetic disk storage, magnetic tape storage, or any other medium capable of carrying or storing data which can be read by a computer.

[0130] It should also be noted that the terms "comprising", "containing", or any other variant thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or apparatus that comprises a list of elements does not include only those elements, but can also include other elements not expressly listed or inherent to such process, method, article or apparatus. Without more limitations, an element defined by the phrase "comprising a" does not exclude the existence of additional identical elements in the process, method, article or apparatus that includes the element.

Claims

1. A power transmission line loss optimization method based on digital twinning, characterized by, The optimization method comprises the following steps: a digital twin model of the power transmission line is constructed, and the line topology and a state vector are determined through the digital twin model; an energy consumption interdependence graph of the power transmission line is constructed through the line topology and the state vector, a multi-dimensional feature tensor is determined according to the current coupling relationship between each node in the energy consumption interdependence graph; local hot spot evolution is performed on the energy consumption interdependence graph to obtain a loss diffusion subgraph, and dynamic relaxation verification is performed on the state quantity of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor; the digital twin model is optimized and fed back according to the multi-dimensional feature tensor and the variable coordination factor to obtain a scheduling feedback strategy; a scheduling priority of the power transmission line is determined, a multi-level control instruction set is determined according to the scheduling priority and the scheduling feedback strategy, and then local loss correction is performed on the power transmission line based on the multi-level control instruction set; wherein constructing an energy consumption interdependence graph of the power transmission line through the line topology and the state vector specifically comprises: mapping the state vector to the nodes and edges of the line topology to obtain different attribute tensors; determining an edge weight model according to all attribute tensors and the energy consumption interdependence of the power transmission line; constructing an energy consumption interdependence graph of the power transmission line based on a Bayesian graph and the edge weight model; wherein the energy consumption interdependence graph is a graph model that integrates topology structure, electrical parameters and energy consumption characteristics, and is used to describe the energy transmission path and coupling strength between each node in the power transmission line under a specific operating state; wherein the local hot spot evolution on the energy consumption interdependence graph specifically comprises: current density monitoring is performed on the energy consumption interdependence graph to obtain a set of hot spot nodes; a diffusion domain is constructed with the hot spot nodes in the set of hot spot nodes as the center, and then a thermal inertia coefficient is extracted from the diffusion domain; a current migration trend is determined based on the thermal inertia coefficient; diffusion evolution is performed on the current migration trend to obtain a loss diffusion subgraph; wherein the local hot spot evolution refers to the identification of hot spot regions in the power transmission line graph and the simulation of thermal diffusion trends to construct a local region model that is highly representative of actual power loss, and the loss diffusion subgraph is a graph model reflecting the thermal coupling relationship between different nodes in the energy transfer process; wherein the dynamic relaxation verification on the state quantity of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor specifically comprises: a time sequence state matrix is constructed based on the loss diffusion subgraph; relaxation boundary calculation is performed on the time sequence state matrix to obtain the tension boundary and the relaxation boundary of the state quantity change of the power transmission line, wherein the tension boundary refers to the maximum response upper limit when the node state quantity appears a continuous rising trend in a continuous time interval, and is used to represent the fluctuation boundary in the state tension accumulation process, and the relaxation boundary refers to the minimum response lower limit when the node state quantity enters a falling trend after the peak value, and is used to depict the response falling range in the state release stage; trend correlation is performed on the tension boundary and the relaxation boundary by difference evolution to obtain the coordinated response fluctuation of the state quantity change of the power transmission line; a variable coordination factor is extracted from the coordinated response fluctuation. 2.The power transmission line loss optimization method based on digital twinning of claim 1, wherein, The digital twin model comprises a twin perception layer, a coupling modeling layer, and a simulation mapping layer, wherein: The twin perception layer is configured to collect electrical operation parameters, structural parameters, and environmental meteorological parameters of the power transmission line to obtain multi-source heterogeneous monitoring data and line topology; The coupling modeling layer is configured to model the multi-source heterogeneous monitoring data to obtain a state vector of physical parameters of the power transmission line; The simulation mapping layer is configured to generate a heat loss estimation surface based on the state vector, and the heat loss estimation surface is configured to determine the energy consumption interdependence relationship of the power transmission line.

3. The digital-twin-based transmission line loss optimization method of claim 1, wherein, Determining the line topology and the state vector through the digital twin model specifically comprises: Determining the line topology through the twin perception layer of the digital twin model; Outputting the state vector through the coupling modeling layer of the digital twin model.

4. The digital-twin-based transmission line loss optimization method of claim 1, wherein, Determining the multi-dimensional feature tensor based on the current coupling relationship between each node in the energy consumption interdependence graph specifically comprises: Extracting an initial state tensor corresponding to each node from the energy consumption interdependence graph; Obtaining the current coupling strength between all nodes in the energy consumption interdependence graph, and then constructing a current influence matrix based on the current coupling strength between all nodes; Supplementing the initial state tensor corresponding to each node through the current influence matrix, and then obtaining the multi-dimensional feature tensor.

5. The digital-twin-based transmission line loss optimization method of claim 1, wherein, Determining the scheduling priority of the power transmission line specifically comprises: Obtaining historical fault records and real-time state quantities of the power transmission line; Constructing a multi-factor risk assessment model based on the historical fault records; Performing risk assessment on the real-time state quantities through the multi-factor risk assessment model to obtain the scheduling priority of the power transmission line.

6. The digital-twin-based transmission line loss optimization method of claim 1, wherein, The state quantity is a set of physical quantities representing the real-time operation state of the power transmission line, and the state quantity comprises electrical parameters and thermal parameters.

7. A power transmission line loss optimization system based on digital twinning for performing the power transmission line loss optimization method based on digital twinning according to any one of claims 1 to 6, characterized in that, The optimization system comprises: A line state twin module configured to construct a digital twin model of the power transmission line, and determine the line topology and the state vector through the digital twin model; A feature tensor extraction module configured to construct an energy consumption interdependence graph of the power transmission line through the line topology and the state vector, and determine a multi-dimensional feature tensor based on the current coupling relationship between each node in the energy consumption interdependence graph; A hotspot evolution verification module configured to perform local hotspot evolution on the energy consumption interdependence graph to obtain a loss diffusion subgraph, and perform dynamic relaxation verification on the state quantity of the power transmission line based on the loss diffusion subgraph to obtain a variable coordination factor; A strategy optimization feedback module configured to perform optimization feedback on the digital twin model based on the multi-dimensional feature tensor and the variable coordination factor to obtain a scheduling feedback strategy; A loss correction execution module configured to determine the scheduling priority of the power transmission line, determine a multi-level control instruction set based on the scheduling priority and the scheduling feedback strategy, and then perform local loss correction on the power transmission line based on the multi-level control instruction set.

Images