A Smart Power Grid Optimization Dispatch Method Based on Digital Twin

Abstract

This invention discloses a smart power grid optimization scheduling method based on digital twins, belonging to the field of power grid optimization scheduling technology. The method includes: constructing a dynamic power grid mirror model based on a standardized power grid data matrix, and generating an evolution trajectory prediction report through a quantum magnetic coupling mechanism; establishing a variable correlation network based on the evolution trajectory prediction report, and generating an optimized scheduling instruction set in conjunction with the standardized data matrix; executing the optimized scheduling instruction set to obtain the magnetic potential energy gradient distribution, and generating a topology reconfiguration scheme by identifying the magnetic potential energy change trend; inputting the topology reconfiguration scheme into the dynamic power grid mirror model to generate an electromagnetic field evolution trajectory, and generating a safety scheduling report in conjunction with the magnetic potential energy gradient distribution. This invention encodes power grid magnetic field data into an evolution trajectory prediction report through a quantum-magnetic coupling mechanism, achieving high-precision risk prediction; and dynamically generates a topology reconfiguration scheme based on the electromagnetic energy accumulation trend, improving the efficiency of power grid optimization scheduling.

Description

Technical Field

[0001] This invention relates to the field of power grid optimization scheduling technology, and in particular to a smart power grid optimization scheduling method based on digital twins. Background Technology

[0002] The deepening development of smart grid construction has propelled digital twin technology to become a core support for power system dispatch. Technological evolution focuses on multi-dimensional technological integration: at the data perception level, wide-area monitoring networks built upon standardized systems enable holographic acquisition and efficient processing of grid operating status; at the modeling and simulation level, dynamic mirror modeling technology is gradually replacing static topology analysis, achieving accurate mapping of the real-time operating status of the grid; and at the decision optimization level, the collaborative application of intelligent algorithms and physical models continuously enhances the adaptive capability of dispatch strategies. Current technological development shows a trend of shifting from single data-driven to a multi-physics coupling paradigm. In particular, the collaborative computing architecture of electromagnetic fields, quantum fields, and magnetic potential fields is becoming the technological frontier of the next generation of digital twin systems, providing a theoretical foundation and methodological support for the stable operation of the grid under high-proportion renewable energy integration.

[0003] However, existing technologies still face two major bottlenecks: First, the problem of the separation between physical and information fields is prominent. Traditional digital twins are mostly limited to data visualization and state monitoring, failing to deeply integrate physical evolution mechanisms such as electromagnetic and magnetic potential fields. This leads to prediction models ignoring quantum-scale correlations and having insufficient ability to capture transient risks such as subsynchronous oscillations. Second, reconstruction decisions are subject to lag and subjectivity. Existing topology optimization relies on manual experience rules and offline verification mechanisms, and the response efficiency is difficult to meet the real-time scheduling requirements. Even with the introduction of intelligent algorithms, they are still limited by weight parameters and are prone to cascading failures due to improper parameters. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a smart power grid optimization scheduling method based on digital twins to solve the problems of inaccurate prediction of new energy fluctuations and lag in topology reconfiguration.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution:

[0007] This invention provides a smart power grid optimization scheduling method based on digital twins, which includes:

[0008] Real-time power grid operation status data is collected and preprocessed to obtain a standardized power grid data matrix. A dynamic power grid mirror model is constructed based on the standardized power grid data matrix, and an evolution trajectory prediction report is generated through a quantum magnetic coupling mechanism. A variable correlation network is established based on the evolution trajectory prediction report, and an optimized scheduling instruction set is generated in combination with the standardized data matrix. The optimized scheduling instruction set is executed to obtain the magnetic potential energy gradient distribution, and a topology reconfiguration scheme is generated by identifying the magnetic potential energy change trend. The topology reconfiguration scheme is input into the dynamic power grid mirror model to generate an electromagnetic field evolution trajectory, and a safe scheduling report is generated in combination with the magnetic potential energy gradient distribution.

[0009] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the real-time operating status data of the power grid includes generator active and reactive power, line power flow parameters, magnetic field strength data, node voltage amplitude, and topology switch opening and closing status.

[0010] The preprocessing includes time alignment, dimensional normalization, topological state encoding, and outlier cleaning.

[0011] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the standardized power grid data matrix includes generator output vectors, node electrical quantity matrices, line power flow tensors, and topological adjacency lists.

[0012] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the specific steps for constructing a dynamic power grid mirror model based on a standardized power grid data matrix are as follows:

[0013] A node connection relationship network is constructed based on the topological adjacency list in the standardized power grid data matrix;

[0014] Load the node electrical quantity matrix into the node connection relationship network to obtain the electrical node network, and bind it with the line power flow tensor to generate the power flow transmission network;

[0015] Based on the power flow transmission network, a dynamic power grid mirror model is constructed by dynamically mapping the generator output vector through power balance constraints.

[0016] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the specific steps for generating the evolution trajectory prediction report through a quantum magnetic coupling mechanism are as follows:

[0017] Extract magnetic field strength data from real-time power grid operation status data, and generate magnetic field distribution heat map data through magnetoelectric coupling equation;

[0018] Based on the dynamic power grid mirror model, the magnetic field distribution heat map data is quantum amplitude encoded to generate an evolution trajectory prediction report.

[0019] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the evolution trajectory prediction report includes the new energy fluctuation range, load curve, and risk heat map.

[0020] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the specific steps for establishing a variable correlation network based on evolution trajectory prediction reports and generating an optimized scheduling instruction set by combining a standardized data matrix are as follows.

[0021] Based on the evolutionary trajectory prediction report, the variable association network is obtained through time-varying mutual information entropy;

[0022] Extract the topological adjacency list from the standardized data matrix, and generate a set of topological constraint variable clusters by overlaying the variable association network;

[0023] Power balance optimization is performed on the set of topology constraint variables to generate an optimized scheduling instruction set.

[0024] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the steps of executing the optimization scheduling instruction set to obtain the magnetic potential energy gradient distribution and generating a topology reconfiguration scheme by identifying the magnetic potential energy change trend are as follows:

[0025] Execute an optimized scheduling instruction set in a digital twin environment to obtain an updated power grid operation status dataset;

[0026] Based on the updated real-time power grid operation status data, a magnetic potential energy gradient distribution heat map is generated by calculating the global magnetic potential energy distribution.

[0027] By identifying the magnetic field lines aggregation direction and intensity variation trend in the magnetic potential energy gradient distribution heatmap, a vector diagram of the magnetic potential energy variation trend is obtained.

[0028] A topology reconstruction scheme is generated based on a magnetic potential energy change trend vector diagram and a magnetic field line navigation path algorithm.

[0029] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the step of inputting the topology reconstruction scheme into the dynamic power grid mirror model to generate the electromagnetic field evolution trajectory means inputting the topology reconstruction scheme into the dynamic power grid mirror model, obtaining the spatiotemporal evolution data of the electromagnetic field through the finite difference method in the time domain, and generating the electromagnetic field evolution trajectory by extracting the electric field intensity evolution curve.

[0030] As a preferred embodiment of the intelligent power grid optimization scheduling method based on digital twins described in this invention, the specific steps for generating a safe scheduling report by combining magnetic potential energy gradient distribution are as follows:

[0031] By combining the electromagnetic field evolution trajectory and the magnetic potential energy gradient distribution, a risk intensity distribution map is generated through the field strength gradient dot product operation;

[0032] Based on the risk intensity distribution map, high-risk equipment and safe operation boundaries are marked using edge detection algorithms, and a safety scheduling report is generated.

[0033] The beneficial effects of this invention are as follows: by encoding the power grid magnetic field data into an evolution trajectory prediction report through a quantum magnetic coupling mechanism, it breaks through the limitations of traditional models in predicting the strong randomness of new energy sources and achieves high-precision risk prediction; combined with magnetic potential energy gradient navigation technology, it dynamically generates topology reconfiguration schemes based on the electromagnetic energy accumulation trend, thereby improving the efficiency of power grid optimization and scheduling. Attached Figure Description

[0034] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0035] Figure 1 A flowchart of a smart power grid optimization scheduling method based on digital twins;

[0036] Figure 2 A flowchart for constructing a dynamic power grid mirror model;

[0037] Figure 3 A flowchart for generating an evolutionary trajectory prediction report;

[0038] Figure 4 A flowchart for generating a topology reconfiguration scheme. Detailed Implementation

[0039] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0040] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0041] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0042] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a smart power grid optimization scheduling method based on digital twins, comprising the following steps:

[0043] S1. Collect real-time power grid operation status data and preprocess it to obtain a standardized power grid data matrix;

[0044] S1.1 Real-time power grid operation status data includes generator active and reactive power, line power flow parameters, magnetic field strength data, node voltage amplitude, and topology switch opening and closing status;

[0045] It should be noted that the active and reactive power of a generator refers to the effective power actually output by the generating equipment and the reactive power used to maintain voltage stability, which is used to characterize the power output characteristics of the power source.

[0046] Line power flow parameters include active power transmission value, reactive power transmission value and current phase angle of transmission lines, reflecting the energy distribution status of the power grid;

[0047] Magnetic field strength data is obtained by using a magnetic induction sensor to acquire the magnetic field strength value around the device, which is used to monitor the degree of electromagnetic field energy accumulation.

[0048] The node voltage amplitude record is the effective value of the AC voltage at each connection point of the power grid, indicating the system voltage stability;

[0049] The opening and closing states of the topology switches are described by Boolean values ​​to indicate the on / off status of the circuit breakers and disconnectors, which determines the physical connection relationship of the power grid.

[0050] S1.2 preprocessing includes time alignment, dimensional normalization, topological state encoding, and outlier cleaning.

[0051] It should be noted that time alignment unifies the acquisition timestamps of multiple source devices through a clock synchronization protocol, eliminating data time differences caused by sensor response delays;

[0052] Dimensional normalization converts voltage and current values ​​into per-unit values ​​and power values ​​into percentage values, solving the problem of incomparability of data with different dimensions.

[0053] Topology state coding converts the on / off states of switches into Boolean logic values, establishing a machine-recognizable topology network.

[0054] Outlier cleaning employs sliding window detection and interpolation algorithms to correct outliers and noisy data.

[0055] S2. Construct a dynamic power grid mirror model based on a standardized power grid data matrix, and generate an evolution trajectory prediction report through a quantum magnetic coupling mechanism;

[0056] The S2.1 standardized power grid data matrix includes generator output vectors, node electrical quantity matrices, line power flow tensors, and topological adjacency lists;

[0057] It should be noted that the generator output vector records the real-time data of active and reactive power output of all generators in the network in the form of a column vector, which is used to characterize the power output characteristics of the power source side.

[0058] The node electrical quantity matrix is ​​a two-dimensional matrix. The rows correspond to the grid node numbers, and the columns store the node voltage amplitude, phase angle, and load power data, which are used to describe the electrical state of each connection point of the grid.

[0059] The line power flow tensor adopts a three-dimensional data structure. The first dimension marks the transmission line number, the second dimension records the time series, and the third dimension stores the active power value, reactive power value and current phase angle of the line, which is used to characterize the spatiotemporal distribution characteristics of power grid energy transmission.

[0060] The topological adjacency list defines the physical structure of the power grid in the form of a Boolean matrix. The rows and columns of the matrix correspond to the node numbers, and the element value "1" indicates that there is a direct connection between the nodes and "0" indicates that there is no connection.

[0061] S2.2 Based on the topological adjacency list in the standardized power grid data matrix, a node connection relationship network is constructed;

[0062] It should be noted that the rows and columns in the topological adjacency list correspond to network node numbers, and the number of network nodes is equal to the dimension of the topological adjacency list matrix. Traverse each row and column of the topological adjacency list; when an element's value is the logical value "1", establish an undirected edge in the node connection relationship network connecting the corresponding row index node and column index node. Assign a power grid node number attribute to each node in the node connection relationship network; assign a line physical connection status attribute to each edge, and output the node connection relationship network.

[0063] S2.3 Load the node electrical quantity matrix into the node connection relationship network to obtain the electrical node network, and bind it with the line power flow tensor to generate the power flow transmission network;

[0064] It should be noted that the node connection network includes the set of power grid nodes and the physical connections between nodes. The node electrical quantity matrix is ​​loaded into the node connection network; the node electrical quantity matrix is ​​a two-dimensional array, with row indices corresponding to power grid node numbers and columns storing node voltage amplitude, phase angle, and load power data; each node in the node connection network is assigned corresponding electrical parameters, forming an electrical node network.

[0065] The line power flow tensor adopts a three-dimensional data structure. The first dimension identifies the transmission line number, the second dimension records the time series, and the third dimension stores the active power value, reactive power value, and current phase angle of the line. The line power flow tensor is bound to the electrical node network. The binding operation uses the transmission line as the associated unit, mapping the power flow data in the line power flow tensor to the corresponding connection edge of the electrical node network to generate the power flow transmission network.

[0066] S2.4 is based on the power flow transmission network and constructs a dynamic power grid mirror model by dynamically mapping the generator output vector through power balance constraints.

[0067] It should be noted that the power flow transmission network includes a set of electrically connected nodes and a set of connected edges with power flow parameters. The load power values ​​of the electrically connected nodes and the active power loss values ​​of the connected edges with power flow parameters are read from the power flow transmission network; the total active power demand is the sum of the load power values ​​of the electrically connected nodes and the active power loss values ​​of the connected edges with power flow parameters.

[0068] The generator output vector stores the total active power output of the generators. It obtains the difference between the total active power output and the total active power demand; dynamically adjusts the active power output values ​​of each generator in the generator output vector according to the generator's regulation capability; the adjusted total active power output must satisfy the requirement that the total active power output equals the total active power demand. The updated generator output vector is mapped to the power flow transmission network; the injected power values ​​of power supply nodes in electrically connected nodes are updated. The state of the power flow transmission network is refreshed, and a dynamic power grid mirror model satisfying power balance constraints is output.

[0069] S2.5 extracts magnetic field intensity data from the real-time operation status data of the power grid and generates magnetic field distribution heat map data through the magnetoelectric coupling equation;

[0070] It should be noted that the magnetic field strength data includes the three-dimensional spatial coordinates of the measurement point, the magnetic field strength value, and the magnetic field strength vector components. A rectangular coordinate system covering the entire power grid equipment area is established based on the three-dimensional spatial coordinates of the measurement point, with the origin of the coordinate system set at the center point of the main transformer. The number of spatial grids in the rectangular coordinate system is then counted.

[0071] For each grid point, the magnetoelectric coupling equation is applied, and the product of the magnetic field intensity vector components in the X / Y / Z directions and the free permeability is used as the magnetic induction intensity. The dimensions of the magnetic field distribution heatmap correspond to the number of spatial grids; the element values ​​of the magnetic field distribution heatmap store the magnetic induction intensity values; a mapping relationship is established between the magnetic field distribution heatmap index and the three-dimensional spatial coordinates of the measurement point. The magnetic field distribution heatmap data is output.

[0072] S2.6 uses a dynamic power grid mirror model to encode the magnetic field distribution heatmap data with quantum amplitude and generates an evolution trajectory prediction report.

[0073] It should be noted that the dynamic power grid mirror model includes a set of electrically connected nodes and a set of connected edges with power flow parameters. The magnetic field distribution heatmap data includes spatial coordinate system parameters and a three-dimensional matrix of magnetic induction intensity. The spatial grid point coordinates of the magnetic field distribution heatmap data are matched with the physical location coordinates of the electrically connected nodes in the dynamic power grid mirror model to establish a spatial coordinate mapping relationship. For each successfully matched electrically connected node, a quantum amplitude encoding operation is performed. First, based on the magnetic induction intensity value, a power grid evolution Hamiltonian is defined. The Hamiltonian includes node current amplitude operator terms and connected edge coupling operator terms; the node current amplitude is taken from the electrical parameters of the electrically connected nodes; the connected edge coupling coefficient is taken from the active power value of the connected edge with power flow parameters. The Hamiltonian is applied to perform unitary evolution on the initial quantum state, and projection measurement is performed on the evolved quantum state to measure the probability distribution in the generator subspace; the high confidence interval is extracted as the new energy fluctuation interval; the expected value is extracted in the load node subspace to generate a load curve time series; the risk probability value is calculated in the equipment risk subspace; and the risk probability value is rendered on the spatial coordinates of the dynamic power grid mirror model to form a risk heatmap. Output an evolution trajectory prediction report. The expression for calculating the risk probability value is:

[0074] ;

[0075] in, This represents the risk probability value. is the base of the exponential function. The curvature coefficient, The magnetic potential energy gradient vector. The electric field intensity gradient vector, This is the risk threshold.

[0076] The range of values ​​for the risk threshold is: Based on the peak value of the magnetic potential energy gradient and the dispersion of the electric field gradient, the boundary conditions for risk triggering are defined.

[0077] The curvature coefficient is determined based on the safety margin of the high-voltage equipment, and its range is as follows: For example, when the safety margin of high-voltage equipment is 10, the curvature coefficient is 0.95.

[0078] S3. Establish a variable correlation network based on the evolution trajectory prediction report, and generate an optimized scheduling instruction set by combining the standardized data matrix;

[0079] The S3.1 evolution trajectory prediction report includes the new energy fluctuation range, load curve, and risk heat map;

[0080] It should be noted that the new energy fluctuation range in the evolution trajectory prediction report refers to the upper and lower limits of the range of variation of the output power of renewable energy power generation equipment. It is used to quantify the uncertainty range of clean energy output such as wind power and photovoltaics, and to provide power supply side fluctuation boundary constraints for dispatching decisions.

[0081] The load curve is a continuous numerical sequence of power consumption at each node in the power grid over time. It reflects the distribution pattern of power demand in different time periods and supports load allocation and power flow control.

[0082] A risk heat map is a matrix distribution map that marks the probability values ​​of equipment operation risks on the geographic spatial coordinates of the power grid. It intuitively displays the spatial distribution of potential hazards such as electromagnetic overload and voltage instability through color depth, and identifies weak links in safety.

[0083] S3.2 Based on the evolutionary trajectory prediction report, the variable association network is obtained through time-varying mutual information entropy;

[0084] It should be noted that the renewable energy output values ​​of each renewable energy power station node within the renewable energy fluctuation range are extracted; the load power values ​​of each load node in the load curve are read; the risk intensity value of each grid in the risk heat map is read, and the total risk intensity value is obtained by accumulating the risk intensity values ​​of each grid. The ratio of the risk intensity value to the total risk intensity value is used as the average risk intensity value, and the equipment in the grid with a risk intensity value higher than the average risk intensity value is marked as high-risk equipment. The renewable energy output value, load power value, and risk intensity value are superimposed to obtain the state interval.

[0085] The total number of data points within the statistical time window and the frequency of the renewable energy output value falling into the state interval are used as the renewable energy output state probability value to obtain the renewable energy output state probability distribution.

[0086] The total number of data points within a statistical time window and the frequency of load power values ​​falling into the state interval are calculated. The ratio of the frequency to the total number of data points within the time window is used as the load power state probability value to obtain the load power state probability distribution.

[0087] The total number of data points within the statistical time window is counted, and the frequency of the new energy output value and load power value falling into the state interval at the same time is obtained; the ratio of the frequency to the total number of data points within the time window is used as the joint probability value of the new energy output state and the load power state, and the joint probability distribution of new energy output and load power is obtained.

[0088] The mutual information entropy value is obtained by performing mutual information entropy calculation based on the probability values ​​of renewable energy output status, load power status, and the joint probability value of renewable energy output status and load power status. The expression for calculating the mutual information entropy value is as follows:

[0089] ;

[0090] in, The mutual information entropy value, This refers to the specific discrete states in the state space of new energy power output. The potential for contributing to new energy sources This refers to the specific discrete states in the load power state space. For load power state space, This represents the joint probability value of the new energy output status and the load power status. Contributing to new energy is in progress The probability value, The load power is in a state The probability value, For The logarithmic function with base quantizes the mutual information entropy value into binary bit units, directly reflecting the minimum amount of information contained in the correlation between the output status of new energy sources and the power status of load.

[0091] The network node set includes new energy power plant nodes, load nodes, and high-risk equipment nodes. The edge set includes connections between new energy power plant nodes and load nodes, connections between new energy power plant nodes and high-risk equipment nodes, and connections between load nodes and high-risk equipment nodes. The network node set and edge set are superimposed to generate a variable association network.

[0092] S3.3 Extracts the topological adjacency list from the standardized data matrix and generates a set of topological constraint variable clusters by overlaying the variable association network;

[0093] It should be noted that, firstly, a node set union operation is performed, which combines the power grid node set, the new energy power station node set, the load node set, and the high-risk equipment node set to form an extended node set; the extended node set includes the power grid node number, the new energy power station entity, the load node entity, and the high-risk equipment entity.

[0094] Then, an edge set merging operation is performed to form an extended edge set by combining the connection edges between new energy power station nodes and load nodes, the connection edges between new energy power station nodes and high-risk equipment nodes, and the connection edges between load nodes and high-risk equipment nodes. The elements of the extended edge set are combinations of node pair numbers. The extended edge set includes two types: physical connection edges and associated edges.

[0095] Starting from any unvisited node on the extended node set and extended edge set, traverse all reachable nodes along the extended edge set; connect all reachable nodes and record them as a cluster. Next, perform entity type integrity verification to check whether each cluster contains four types of elements: power grid node number, new energy station entity, load node entity, and high-risk equipment entity; if any entity type is missing, merge the cluster with the adjacent cluster.

[0096] Physical connection constraint strengthening is implemented, and the existence of physical connection edges defined in the topological adjacency list within the cluster is checked. If no physical connection edges exist within the cluster, the cluster is decomposed into single-node clusters. Finally, a set of topological constraint variable clusters is generated.

[0097] S3.4 performs power balance optimization on the set of topology constraint variables and generates an optimized scheduling instruction set.

[0098] It should be noted that the median of the new energy fluctuation range is used as the reference output; the sum of the squares of the deviations between the new energy output status and the reference output, plus the total risk intensity value of high-risk equipment, is used as the constraint condition.

[0099] The objective function and constraints are transformed into a standard quadratic programming form; the target output value of the new energy power station entity is output as the optimization result; and the risk control action intensity of high-risk equipment entities is determined.

[0100] For each renewable energy power station entity, an output instruction field is created, containing the power station's unique identifier and target output value. For each high-risk equipment entity, a risk control instruction field is created, containing the high-risk equipment's unique identifier and the intensity of the risk control action. Cluster instructions are encapsulated: each cluster outputs an independent instruction unit; the instruction unit contains a cluster identifier, a list of renewable energy output instructions, and a list of risk control instructions. The cluster instruction units are then aggregated to form an optimized scheduling instruction set.

[0101] S4. Execute the optimized scheduling instruction set to obtain the magnetic potential energy gradient distribution, and generate a topology reconstruction scheme by identifying the trend of magnetic potential energy change.

[0102] S4.1 Executes an optimized scheduling instruction set in the digital twin environment to obtain an updated power grid operation status dataset;

[0103] It should be noted that the process involves traversing the list of new energy power output instructions, extracting the unique identifier and target output value of the new energy power station from the instruction fields, locating the corresponding new energy power station entity in the dynamic power grid mirror model, and adjusting the output value of the new energy power station entity to the target output value.

[0104] Traverse the list of risk control instructions for high-risk equipment; extract the unique identifier and control action description of the high-risk equipment from the instruction fields; locate the corresponding high-risk equipment entity in the dynamic power grid mirror model; and execute load reduction according to the control action description field.

[0105] Refresh the state parameters of the dynamic power grid mirror model; deploy sensors to collect generator active and reactive power values, line power flow parameter values, node voltage amplitude, and magnetic field strength data. Encapsulate the updated data; generate an updated power grid operating status dataset; the updated power grid operating status dataset includes generator active and reactive power fields, line power flow parameter fields, magnetic field strength data fields, and node voltage amplitude fields.

[0106] S4.2 generates a magnetic potential energy gradient distribution heatmap by calculating the global magnetic potential energy distribution based on the updated real-time power grid operation status data;

[0107] It should be noted that a spatial grid coordinate system is established; the spatial grid coordinate system covers the entire area of ​​the power grid equipment, and the number of spatial grids is counted; the origin of the spatial grid coordinate system is aligned with the origin of the spatial coordinate system of the dynamic power grid mirror model to obtain the spatial grid coordinates.

[0108] Read the magnetic field intensity vector components of the spatial grid; apply the magnetoelectric coupling equation to take the product of the vacuum permeability and the magnetic field intensity vector components as the magnetic induction intensity components.

[0109] The magnetic potential energy density is defined as the ratio of the sum of squares of the magnetic induction intensity components to twice the free permeability. A magnetic potential energy density distribution matrix is ​​filled, with its dimensions corresponding to the number of spatial grid cells; the matrix element values ​​store the magnetic potential energy density values; and the matrix row and column indices map the spatial grid coordinates.

[0110] Spatial gradient calculations are performed on the magnetic potential energy density distribution matrix. Using the physical location spacing of the magnetic field sensors as the grid spacing, the central difference method is used along the X-axis of the spatial grid coordinate system to obtain the partial derivative component of the ratio of the difference in magnetic potential energy density values ​​between adjacent grid points to twice the grid spacing. Similarly, along the Y-axis and Z-axis of the spatial grid coordinate system, the central difference method is used to obtain the partial derivative component of the ratio of the difference in magnetic potential energy density values ​​between adjacent grid points to twice the grid spacing. The gradient vector is then synthesized, containing partial derivative components in the X, Y, and Z directions.

[0111] S4.3 By identifying the magnetic field line aggregation direction and intensity variation trend in the magnetic potential energy gradient distribution heatmap, a vector diagram of the magnetic potential energy variation trend is obtained.

[0112] It should be noted that the magnetic potential energy gradient distribution heatmap includes a gradient intensity matrix and a gradient direction matrix. Local maxima in the gradient intensity matrix are selected as seed points; based on the gradient direction angular components recorded in the gradient direction matrix; starting from the local maxima in the gradient intensity matrix, the process iteratively advances along the vector direction recorded in the gradient direction matrix, recording the path point coordinate sequence and connecting them to form a continuous magnetic field line path set.

[0113] The spatial grid coordinate system is divided into density calculation grids; the number of magnetic field line paths passing through each density calculation grid is counted; the ratio of the number of magnetic field line paths to the number of density calculation grids is the mean magnetic field line density.

[0114] Scan the magnetic field density distribution; mark regions where the magnetic field density value is greater than the average magnetic field density as clustered regions; extract the direction of the magnetic field path within the clustered regions; calculate the total number of vectors for all magnetic field path directions within the clustered regions, and obtain the total vector value by summing them, using the ratio of the total vector value to the total number of vectors as the average vector value; the direction where the vector of the magnetic field path direction is greater than the average vector value is taken as the magnetic field clustering direction.

[0115] The gradient intensity value is tracked along the magnetic field line path; the intensity change rate is obtained based on the ratio of the difference in gradient intensity value between adjacent grid points to the grid spacing; the cumulative change is obtained along the intensity change rate of the magnetic field line path.

[0116] Create a vector matrix, where the dimensions of the vector matrix correspond to the number of spatial grids; matrix elements store vector structures; vector structures contain directional and intensity components; intensity components record the cumulative change values, ultimately generating a vector map of magnetic potential energy change trends.

[0117] S4.4 generates a topology reconstruction scheme based on the magnetic potential energy change trend vector diagram and the magnetic field line navigation path algorithm.

[0118] It should be noted that a set of nodes for the power grid topology graph is created; the elements of the node set are the power grid node numbers; the power grid node numbers are derived from the row and column indices of the topology adjacency list. A set of edges for the power grid topology graph is created; each element of the topology adjacency list is traversed; when the element value is the logical value "1", the node pair with the corresponding row and column index is added to the edge set; each edge is assigned the initial connection status attribute "connecting".

[0119] Traverse each edge in the power grid topology graph edge set; obtain the spatial coordinates of the nodes at both ends of the edge; the spatial coordinates are derived from the spatial mapping relationship of the dynamic power grid mirror model. Locate the corresponding position in the magnetic potential energy change trend vector diagram; read the intensity component values ​​of the vector structure at the position; use the arithmetic mean of the intensity component values ​​of the nodes at both ends as the basic cost of the edge.

[0120] In the power grid topology diagram, nodes in renewable energy-rich areas are selected as path source points; these nodes are determined by the peak values ​​of renewable energy fluctuation intervals in the evolution trajectory prediction report. Nodes in load center areas are selected as path sink points; these nodes are determined by the peak nodes of the load curve in the evolution trajectory prediction report. All nodes in the power grid topology diagram are traversed, and all path sink points of the path source points are identified to obtain the total cost of adjacent edges. The sum of the cost of the path source point and the total cost of adjacent edges is the path cost of the path sink point. If the path cost of the path sink point is less than its existing path cost, the path source point is recorded as the predecessor node of the path sink point, generating the optimal path node sequence.

[0121] Detect adjacent node pairs in the optimal path node sequence; query the node pair connection status in the topology adjacency table; if the connection status is logic value "0"; generate a closing switch instruction; the instruction includes a unique switch identifier and the action type "closing". Traverse all connected edges in the power grid topology graph; if the total edge value is greater than the average risk intensity, generate a disconnecting switch instruction; the instruction includes a unique switch identifier and the action type "disconnecting".

[0122] Encapsulate a list of switch operation instructions; each element in the switch operation instruction list includes a switch identifier field and an action type field; the action type field takes the value "closed" or "open". Output a topology reconfiguration scheme; the topology reconfiguration scheme is stored as a structured instruction set.

[0123] S5. Input the topology reconfiguration scheme into the dynamic power grid mirror model to generate the electromagnetic field evolution trajectory, and combine it with the magnetic potential energy gradient distribution to generate a safe dispatch report.

[0124] S5.1 inputs the topology reconfiguration scheme into the dynamic power grid mirror model, obtains the spatiotemporal evolution data of the electromagnetic field through the finite-difference time-domain method, and generates the electromagnetic field evolution trajectory by extracting the electric field intensity evolution curve.

[0125] It should be noted that the topology reconfiguration scheme is loaded into the dynamic power grid mirror model; the topology reconfiguration scheme includes a list of switch operation instructions. The switch status update operation is performed by iterating through each instruction in the switch operation instruction list; the switch unique identifier field is read; and the switch status is updated according to the action type field: if the action type is "closed", the switch connection status is set to logic value "1"; if the action type is "open", the switch connection status is set to logic value "0".

[0126] The spatial domain of the dynamic power grid mirror model is divided into a cubic grid; the grid node coordinates are strictly aligned with the device positions in the dynamic power grid mirror model.

[0127] When the electric field strength component and the magnetic field strength component are zero, the magnetic field strength component is updated according to Faraday's law. The change of magnetic field strength with time is excited by the spatial rotation of electric field strength. The spatial derivative is approximated by the central difference method. The symmetrical difference of electric field strength values ​​between adjacent grid points is taken to obtain the approximate rate of change of electric field strength component along the coordinate axis.

[0128] The electric field intensity component is updated according to Ampere's law, and the change of electric field intensity over time is excited by the spatial rotation of the magnetic field intensity. The spatial derivative is approximated using the central difference method, and the symmetrical difference between the magnetic field intensity values ​​of adjacent grid points is taken to obtain an approximate value of the rate of change of the magnetic field intensity component along the coordinate axis. An exponential decay function is applied to process the magnetic field intensity component and the electric field intensity component, and the spatiotemporal evolution data of the electromagnetic field are output.

[0129] Monitoring points are set up at the physical locations of new energy power plants, load nodes, and core components of high-risk equipment. The electric field intensity components at each monitoring point are recorded at each time step. A two-dimensional array of time and electric field intensity is created, with the time series as the horizontal axis and the electric field intensity components as the vertical axis, generating an evolution curve of the electric field intensity at each monitoring point. The evolution curve of the electric field intensity at each monitoring point is then encapsulated to generate an electromagnetic field evolution trajectory.

[0130] S5.2 Combines the electromagnetic field evolution trajectory with the magnetic potential energy gradient distribution, and generates a risk intensity distribution map through field strength gradient dot product operation;

[0131] It should be noted that the process involves traversing each grid point in the spatial grid coordinate system and reading the electric field intensity component values ​​from the spatiotemporal evolution data of the electromagnetic field. Along the three axes of the spatial grid coordinate system, the differences in electric field intensity component values ​​between adjacent grid points are obtained; the ratio of twice the grid spacing to the difference in electric field intensity component values ​​between adjacent grid points is the spatial rate of change component. The resulting electric field spatial rate of change vector is then synthesized.

[0132] Create a risk intensity distribution matrix; the dimensions of the risk intensity distribution matrix are consistent with the spatial grid coordinate system; the matrix element values ​​store the risk intensity values. The matrix row and column indices map to the spatial grid coordinates; write the risk intensity value of each grid point into the generated risk intensity distribution map.

[0133] S5.3 uses an edge detection algorithm to mark high-risk equipment and safe operation boundaries based on the risk intensity distribution map, and generates a safety scheduling report.

[0134] It should be noted that the matrix elements of the risk intensity distribution map store the risk intensity values; the matrix row and column indices map spatial grid coordinates. An edge detection algorithm is executed, using the Sobel operator to calculate the spatial gradient magnitude of the risk intensity distribution matrix; local maxima are preserved along the gradient direction; gradient magnitude peaks are marked as strong edge points; gradient magnitude valleys are marked as weak edge points; and strong edge points are connected to adjacent weak edge points to form continuous edge lines. The expression for calculating the spatial gradient magnitude of the risk intensity is:

[0135] ;

[0136] in, This represents the spatial gradient magnitude of risk intensity. Risk intensity along spatial direction The partial derivatives, This is the risk intensity distribution matrix. Let x be the horizontal axis of the spatial rectangular coordinate system. Risk intensity along spatial direction The partial derivatives, y is the vertical axis of the spatial rectangular coordinate system.

[0137] Locate the contour region within a continuous edge line; areas within the contour region with risk intensity values ​​higher than the average risk intensity value are marked as high-risk areas. Match the spatial coordinates of the high-risk areas in the dynamic power grid mirror model; obtain the equipment entities covered by the high-risk areas; equipment entities include renewable energy power plant entities, load node entities, and high-risk equipment entities. Label the equipment entities as high-risk equipment; record the equipment's unique identifier and peak risk intensity value.

[0138] Create a high-risk device list; the high-risk device list elements include a device type field, a device identifier field, and a risk intensity peak field. Encapsulate a security scheduling report; the security scheduling report includes a high-risk device list field and a risk intensity distribution map identifier field.

[0139] This embodiment also provides a computer device applicable to the intelligent power grid optimization scheduling method based on digital twins, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the intelligent power grid optimization scheduling method based on digital twins as proposed in the above embodiment.

[0140] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0141] This embodiment also provides a storage medium storing a computer program. When executed by a processor, the program implements the intelligent power grid optimization scheduling method based on digital twins as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0142] In summary, this invention achieves high-precision risk prediction by encoding power grid magnetic field data into an evolution trajectory prediction report through a quantum magnetic coupling mechanism, overcoming the limitations of traditional models in predicting the strong randomness of new energy sources; and by combining magnetic potential energy gradient navigation technology to dynamically generate topology reconfiguration schemes based on electromagnetic energy accumulation trends, thereby improving the efficiency of power grid optimization and scheduling.

[0143] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A smart power grid optimization scheduling method based on digital twins, characterized in that: include, Collect real-time power grid operation status data and preprocess it to obtain a standardized power grid data matrix; A dynamic power grid mirror model is constructed based on a standardized power grid data matrix, and an evolution trajectory prediction report is generated through a quantum magnetic coupling mechanism. The specific steps are as follows. A node connection relationship network is constructed based on the topological adjacency list in the standardized power grid data matrix; Load the node electrical quantity matrix into the node connection relationship network to obtain the electrical node network, and bind it with the line power flow tensor to generate the power flow transmission network; Based on the power flow transmission network, a dynamic power grid mirror model is constructed by dynamically mapping the generator output vector through power balance constraints. Extract magnetic field strength data from real-time power grid operation status data, and generate magnetic field distribution heat map data through magnetoelectric coupling equation; Based on the dynamic power grid mirror model, the magnetic field distribution heat map data is quantum amplitude encoded to generate an evolution trajectory prediction report; A variable correlation network is established based on the evolution trajectory prediction report, and an optimized scheduling instruction set is generated by combining it with a standardized data matrix; Execute the optimized scheduling instruction set to obtain the magnetic potential energy gradient distribution, and generate a topology reconstruction scheme by identifying the trend of magnetic potential energy change. The topology reconfiguration scheme is input into the dynamic power grid mirror model to generate the electromagnetic field evolution trajectory, and combined with the magnetic potential energy gradient distribution to generate a safe dispatch report.

2. The intelligent power grid optimization scheduling method based on digital twins according to claim 1, characterized in that: The real-time operating status data of the power grid includes generator active and reactive power, line power flow parameters, magnetic field strength data, node voltage amplitude, and topology switch opening and closing status. The preprocessing includes time alignment, dimensional normalization, topological state encoding, and outlier cleaning.

3. The intelligent power grid optimization scheduling method based on digital twins according to claim 2, characterized in that: The standardized power grid data matrix includes generator output vectors, node electrical quantity matrices, line power flow tensors, and topological adjacency lists.

4. The intelligent power grid optimization scheduling method based on digital twins according to claim 3, characterized in that: The evolution trajectory prediction report includes the new energy fluctuation range, load curve, and risk heat map.

5. The intelligent power grid optimization scheduling method based on digital twins according to claim 4, characterized in that: The specific steps for establishing a variable correlation network based on the evolutionary trajectory prediction report and generating an optimized scheduling instruction set by combining it with a standardized data matrix are as follows. Based on the evolutionary trajectory prediction report, the variable association network is obtained through time-varying mutual information entropy; Extract the topological adjacency list from the standardized data matrix, and generate a set of topological constraint variable clusters by overlaying variable association networks; Power balance optimization is performed on the set of topology constraint variables to generate an optimized scheduling instruction set.

6. The intelligent power grid optimization scheduling method based on digital twins according to claim 5, characterized in that: The execution of the optimized scheduling instruction set obtains the magnetic potential energy gradient distribution, and generates a topology reconstruction scheme by identifying the trend of magnetic potential energy changes. The specific steps are as follows. Execute an optimized scheduling instruction set in a digital twin environment to obtain an updated power grid operation status dataset; Based on the updated real-time power grid operation status dataset, a magnetic potential energy gradient distribution heatmap is generated by calculating the global magnetic potential energy distribution. By identifying the magnetic field lines aggregation direction and intensity variation trend in the magnetic potential energy gradient distribution heatmap, a vector diagram of the magnetic potential energy variation trend is obtained. A topology reconstruction scheme is generated based on a magnetic potential energy change trend vector diagram and a magnetic field line navigation path algorithm.

7. The intelligent power grid optimization scheduling method based on digital twins according to claim 6, characterized in that: The process of inputting the topology reconstruction scheme into the dynamic power grid mirror model to generate the electromagnetic field evolution trajectory refers to inputting the topology reconstruction scheme into the dynamic power grid mirror model, obtaining the spatiotemporal evolution data of the electromagnetic field through the finite-difference time-domain method, and generating the electromagnetic field evolution trajectory by extracting the electric field intensity evolution curve.

8. The intelligent power grid optimization scheduling method based on digital twins according to claim 7, characterized in that: The specific steps for generating a safe scheduling report by combining the magnetic potential energy gradient distribution are as follows. By combining the electromagnetic field evolution trajectory and the magnetic potential energy gradient distribution, a risk intensity distribution map is generated through the field strength gradient dot product operation; Based on the risk intensity distribution map, high-risk equipment and safe operation boundaries are marked using edge detection algorithms, and a safety scheduling report is generated.

Images