Complex power grid controllable phase shifter site selection method and system based on dynamic power flow trend prediction

Abstract

The invention relates to the technical field of power system optimization and power flow control, in particular to a complex power grid controllable phase shifter site selection method and system based on dynamic power flow trend prediction, and the method comprises the steps: setting a full time domain dimension, obtaining power grid power flow data and phase shifter parameters, and analyzing and predicting the dynamic power flow distribution of a power grid based on the power grid power flow data; establishing a power load path according to the dynamic power flow distribution, evaluating the importance of the path, and generating an important path evaluation result; combining the phase shifter parameters and the important path evaluation result to construct a phase shifter site selection multi-objective optimization model, and generating a plurality of phase shifter site selection schemes; and carrying out multi-scene simulation verification on the phase shifter site selection scheme, verifying the applicability of the phase shifter site selection scheme in different scenes, and screening to obtain a phase shifter site selection result. According to the method, the problems that in an existing phase shifter site selection method, the dynamic change trend of power grid power flow is not considered, critical path evaluation is insufficient, and multi-scene adaptability verification is lacked are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of power system optimization and power flow control, and particularly to a method and system for locating a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction. Background Art

[0002] Currently, in order to solve the problem of easy overload of lines during the operation of the power grid, phase shifters, as an economical power flow control device, are widely used in modern power systems. By adjusting the voltage phase angle, phase shifters can optimize the power flow distribution, relieve problems such as line overload and power flow congestion, thereby improving the stability and economy of the power grid. However, the adjustment effect of the phase shifter highly depends on the scientificity and accuracy of its installation position.

[0003] Existing methods for locating phase shifters are mainly divided into two categories: based on sensitivity analysis and based on optimization models. The former screens the installation positions by analyzing the sensitivity of the phase shifter to the power grid power flow. The calculation complexity is low, but it is limited to the current power flow information and does not consider the dynamic changes of the power flow. The latter constructs a multi-objective optimization model and combines algorithms to solve the location of the phase shifter. However, the input data is mostly based on the real-time working conditions and fails to effectively utilize historical data and future power flow trends. The existing technologies fail to consider the power flow change trend in the full time domain, only locate based on the current working conditions, ignore the impact of historical and future power flow changes on the operation of the power grid, and cannot accurately identify the key paths prone to overload or congestion in the power grid, which affects the scientificity of the location. Moreover, there is a lack of multi-scenario verification. The existing location schemes have not been fully tested in typical scenarios such as high load, fault, or new energy access, resulting in poor adaptability and reliability of the schemes. These methods generally have problems of insufficient long-term adaptability to the operation of the power grid.

[0004] The information disclosed in this background art section is only intended to deepen the understanding of the overall background art of the present disclosure, and should not be regarded as an admission or any form of suggestion that this information constitutes the prior art known to those skilled in the art. Summary of the Invention

[0005] The present invention provides a method and system for locating a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction, which can effectively solve the problems in the background art.

[0006] In order to achieve the above object, the technical solution adopted by the present invention is:

[0007] A method for locating a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction, the method comprising:

[0008] Set the full time domain dimension, obtain the power grid power flow data and phase shifter parameters according to the full time domain dimension, and predict the dynamic power flow distribution of the power grid based on the power grid power flow data;

[0009] Establish a power load path based on the dynamic power flow distribution, evaluate the path importance of the power load path, and generate an important path evaluation result;

[0010] Combine the phase shifter parameters and the important path evaluation result to construct a multi-objective optimization model for phase shifter location selection, and generate several phase shifter location selection schemes;

[0011] Conduct multi-scenario simulation verification on several phase shifter location selection schemes, verify the applicability of the phase shifter location selection schemes under different power flow scenarios, and screen to obtain the phase shifter location selection result.

[0012] Furthermore, analyze the power grid power flow data to predict the dynamic power flow distribution of the power grid, including:

[0013] The full time domain dimension includes three dimensions: historical state, real-time state, and future state;

[0014] Integrate the power grid power flow data according to the full time domain dimension to construct a full time domain power grid power flow data set;

[0015] Establish a full time domain load prediction model, and based on the full time domain power grid power flow data set, predict the future state of the current power line load to generate a line load rate prediction value and a line sensitivity prediction value;

[0016] Compare the line load rate prediction value and the line sensitivity prediction value with the power grid power flow data according to the change of time state to predict the dynamic power flow distribution.

[0017] Furthermore, predict the future state of the current power line load, including:

[0018] Use the historical state power grid power flow data as the training set and the real-time state power grid power flow data as the verification set to train and generate the full time domain load prediction model;

[0019] Analyze the real-time state power grid power flow data, obtain the ratio of the line power flow value to the maximum allowable capacity of the line, and calculate the load margin based on the ratio of the line power flow value to the maximum allowable capacity of the line;

[0020] The full time domain load prediction model calculates the line load rate prediction value based on the ratio of the line power flow value to the maximum allowable capacity of the line;

[0021] Based on the full time domain load prediction model, establish a dynamic correlation between the full time domain power flow change between lines and the load margin, and generate the line sensitivity prediction value of each line in the power network according to the dynamic correlation.

[0022] Furthermore, establish a power load path according to the dynamic power flow distribution, including:

[0023] Analyze the power flow distribution characteristics of the power grid operation based on the dynamic power flow distribution in the full time domain dimension, and identify the convergence source nodes and divergence source nodes of the power flow;

[0024] Analyze the power flow transmission relationship between the convergence source nodes and the divergence source nodes, and define the power load path construction rules with the goal of minimizing the path power flow loss and avoiding path overload based on the power flow transmission relationship;

[0025] Based on the power load path construction rules, establish the topological structure of the power load path from the divergence source nodes to the convergence source nodes;

[0026] Verify the load adaptability of the established power load path, and update the topological structure of the power load path according to the verification results.

[0027] Furthermore, evaluate the path importance of the power load path, including:

[0028] Based on the power flow distribution characteristics, perform weighted accumulation on each power load path, and calculate the comprehensive power flow carrying capacity of each power load path;

[0029] Combined with the power grid power flow data and the comprehensive power flow carrying capacity in the full time domain dimension, analyze the operation stability of the power load path, and calculate the operation reliability coefficient;

[0030] Based on the topological structure of the power load path and the dynamic power flow distribution, simulate the impact of key line faults in the power load path on the overall power flow change of the power grid, and calculate the fault sensitivity coefficient;

[0031] Generate a path importance index according to the operation reliability coefficient and the fault sensitivity coefficient, and sort the path importance indexes of each power load path to generate the important path evaluation result.

[0032] Furthermore, generate several shunt phase modifier location selection schemes, including:

[0033] Based on the dynamic power flow distribution and the shunt phase modifier parameters, calculate the shunt phase modifier sensitivity coefficient of each power load path, and construct a shunt phase modifier sensitivity matrix according to the shunt phase modifier sensitivity coefficient;

[0034] Combined with the important path evaluation result and the shunt phase modifier sensitivity matrix, analyze the adjustment effect of the shunt phase modifier on each power load path, and construct a shunt phase modifier candidate line set;

[0035] Based on the dynamic power flow distribution in the full time domain dimension, establish the shunt phase modifier location selection multi-objective optimization model in combination with the shunt phase modifier sensitivity matrix;

[0036] Parse the candidate line set of the phase shifter according to the multi-objective optimization model for the phase shifter location selection, and generate a number of phase shifter location selection schemes.

[0037] Further, calculate the phase shifter sensitivity coefficients of each of the power load paths, including:

[0038] Based on the dynamic power flow distribution in the full time domain dimension, obtain the basic operation data of each line in the power load path;

[0039] Analyze the power flow regulation characteristics of each of the power load paths based on the basic operation data and the phase shifter parameters;

[0040] According to the power flow regulation characteristics, calculate the influence amount of the power flow change on other power load paths when a phase shifter is installed on each of the power load paths;

[0041] Perform a weighted calculation on the influence amount of the power flow change, and comprehensively consider the phase shifter parameters, and generate the phase shifter sensitivity coefficients corresponding to the power load paths according to the calculation weights of the influence amount of the power flow change of the power load paths.

[0042] Further, perform multi-scenario simulation verification on a number of the phase shifter location selection schemes, including:

[0043] Based on the dynamic power flow distribution and the historical power grid power flow data, construct simulation scenarios covering various operating states;

[0044] According to the phase shifter parameters, the power grid topology structure and the power flow distribution characteristics, set the simulation operation parameters required for the simulation scenarios;

[0045] Perform power flow calculations for each of the phase shifter location selection schemes in different simulation scenarios according to the simulation operation parameters;

[0046] Based on the results of the power flow calculations, evaluate the power flow regulation capabilities of each of the phase shifter location selection schemes in different simulation scenarios and the impact on the power grid operation stability, and quantify the location adaptability of each of the phase shifter location selection schemes;

[0047] According to the location adaptability, evaluate and rank each of the phase shifter location selection schemes, and screen out the phase shifter location selection schemes that can maintain the operation performance in multiple simulation scenarios, and generate the phase shifter location selection results.

[0048] A controllable phase shifter location selection system for a complex power grid based on dynamic power flow trend prediction, the system includes:

[0049] The analysis and prediction module sets the full-time domain dimension, obtains the power grid power flow data and phase shifter parameters according to the full-time domain dimension, and analyzes and predicts the dynamic power flow distribution of the power grid based on the power grid power flow data;

[0050] The path evaluation module establishes a power load path according to the dynamic power flow distribution, evaluates the path importance of the power load path, and generates an important path evaluation result;

[0051] The scheme generation module combines the phase shifter parameters and the important path evaluation result, constructs a multi-objective optimization model for phase shifter location selection, and generates several phase shifter location selection schemes;

[0052] The verification location selection module conducts multi-scenario simulation verification on several phase shifter location selection schemes, verifies the applicability of the phase shifter location selection scheme under different power flow scenarios, and filters to obtain the phase shifter location selection result.

[0053] Furthermore, the analysis and prediction module includes:

[0054] The time domain definition unit, and the full-time domain dimension includes three dimensions: historical state, real-time state, and future state;

[0055] The information integration unit integrates the power grid power flow data according to the full-time domain dimension to construct a full-time domain power grid power flow data set;

[0056] The model analysis unit establishes a full-time domain load prediction model, predicts the future state of the current power line load based on the full-time domain power grid power flow data, and generates a line load rate prediction value and a line sensitivity prediction value;

[0057] The comparison and prediction unit compares the line load rate prediction value and the line sensitivity prediction value with the power grid power flow data according to the time state change to predict the dynamic power flow distribution.

[0058] Through the technical solution of the present invention, the following technical effects can be achieved:

[0059] It solves the problems in the existing phase shifter location selection method that do not consider the dynamic change trend of the power grid power flow, insufficient key path evaluation, and lack of multi-scenario adaptability verification. Combining the full-time domain power flow data, the phase shifter sensitivity matrix and multi-objective optimization, it generates a scientific location selection scheme that meets the current and future power grid requirements, effectively improving the power flow regulation ability and operation stability of the power grid.

[0060] The above description is only an overview of the technical solution of the present application. In order to be able to understand the technical means of the present application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of the present application more obvious and understandable, the specific embodiments of the present application are specifically given below. Brief Description of the Drawings

[0061] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the accompanying drawings required for the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only some embodiments recorded in the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0062] Figure 1 It is a schematic flowchart of a method for locating a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction;

[0063] Figure 2 It is a schematic structural diagram for obtaining dynamic power flow distribution;

[0064] Figure 3 It is a schematic flowchart for establishing a power load path;

[0065] Figure 4 It is a schematic flowchart for generating a phase shifter location selection scheme;

[0066] Figure 5 It is a schematic structural diagram for multi-scenario verification. Specific embodiments

[0067] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, rather than all embodiments.

[0068] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs. The terms used in the description of the present invention are only for the purpose of describing specific embodiments, and are not intended to limit the present invention. The term "and / or" used herein includes any and all combinations of one or more of the related listed items.

[0069] Embodiment 1;

[0070] As Figure 1 shown, the present application provides a method for locating a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction. The method includes:

[0071] Set the full-time domain dimension, obtain the power grid power flow data and phase shifter parameters according to the full-time domain dimension, and analyze and predict the dynamic power flow distribution of the power grid based on the power grid power flow data;

[0072] Establish a power load path according to the dynamic power flow distribution, evaluate the path importance of the power load path, and generate an important path evaluation result;

[0073] Combined with the phase shifter parameters and the evaluation results of important paths, a multi-objective optimization model for phase shifter location selection is constructed to generate several phase shifter location selection schemes;

[0074] Perform multi-scenario simulation verification on several phase shifter location selection schemes to verify the applicability of the phase shifter location selection schemes under different power flow scenarios, and screen out the phase shifter location selection results.

[0075] Specifically, first set the full-time domain dimension, including three dimensions: historical state, real-time state, and future state. Through power grid and power flow monitoring, obtain the power grid power flow data in the historical state and real-time state, and combined with the phase shifter parameters, construct the full-time domain power flow data set of the power grid. Use the BP neural network algorithm to predict the dynamic power flow distribution in the future state, and generate the predicted values of the line load rate and the "N-1" line sensitivity prediction value for each line; calculate the predicted value of the line load rate according to the ratio of the line power flow value to the maximum allowable capacity, and quantify the overload risk of the line; through the LODF matrix, analyze the power flow transfer characteristics of the line under fault conditions, and obtain the "N-1" line sensitivity prediction value of each line in the power grid, which is used to characterize the sensitivity of the power flow distribution change; based on the prediction results, analyze the change trend of the power flow distribution in the power grid, identify the source nodes and divergence nodes of the power flow convergence, use the Pagerank algorithm to sort the importance of all lines in the power grid, and generate the comprehensive importance index of the line according to the weight coefficients of the predicted value of the line load rate and the "N-1" line sensitivity prediction value, and select the lines with higher importance to form the key path set (L1); according to the power flow distribution characteristics and path transmission rules, analyze the power flow carrying capacity of each line in the path, quantify the power flow transmission effectiveness of the path, calculate the fault sensitivity coefficient of the path by simulating the fault scenarios of the key lines, and comprehensively consider the power flow carrying capacity, operation reliability and fault sensitivity of the path to generate the importance evaluation results of the path, and sort the priorities of the paths to establish the topological structure of the power load path; on the basis of the path evaluation, combined with the phase shifter sensitivity matrix, establish a multi-objective optimization model for phase shifter location selection. The phase shifter sensitivity matrix is used to quantify the adjustment ability of the phase shifter installation on the power flow distribution of the target line and the surrounding lines, solve the model, and generate multiple phase shifter location selection schemes; for the generated phase shifter location selection schemes, construct simulation scenarios covering various operating conditions, including high load, low load, new energy access and line fault scenarios, etc. Under each simulation scenario, perform power flow calculations to verify the applicability of each location selection scheme, and the evaluation indicators include power flow adjustment ability, node voltage deviation, total system loss, and operation stability; according to the multi-scenario simulation verification results, conduct a comprehensive performance evaluation and ranking of the phase shifter location selection schemes, and finally select the optimal location selection scheme that takes into account both economy and reliability as the phase shifter location selection result.

[0076] The technical solution of the present invention solves the problems of not considering the dynamic change trend of power grid flow, insufficient critical path evaluation and lack of multi-scenario adaptability verification in the existing phase shifter site selection method. By combining full time domain power flow data, phase shifter sensitivity matrix and multi-objective optimization, a scientific site selection plan that adapts to current and future power grid needs is generated, effectively improving the power grid's power flow regulation capability and operational stability.

[0077] Further, if Figure 2 As shown, the dynamic power flow distribution of the power grid is predicted based on the power grid power flow data analysis, including:

[0078] The full time domain dimension includes three dimensions: historical state, real-time state and future state;

[0079] Integrate power grid flow data according to the full time domain dimension and construct a full time domain power grid flow data set;

[0080] Establish a full-time load prediction model, predict the future state of the current power line load based on the full-time power grid flow data, and generate line load rate prediction values ​​and line sensitivity prediction values;

[0081] The predicted values ​​of line load rate and line sensitivity are compared with the grid power flow data with time-dependent changes to predict the dynamic power flow distribution.

[0082] As a preferred embodiment of the above, a full time domain dimension is constructed, and the historical state, real-time state and future state of the power grid are used as the three core time states for analysis. The historical state includes the collection of power grid flow operation data within a certain period of time in the past, such as line load, node voltage and fault events, which are used as the training set of the prediction model; the real-time state obtains the current flow operation data through the power grid operation monitoring system as the model input and verification; the future state is based on the flow data of the historical state and the real-time state, and the flow distribution within a certain time range in the future is speculated through the prediction model, so as to construct a full time domain analysis framework; the above historical state and real-time state data are integrated to form a full time domain flow data set. In the process of data integration, the collected data is preprocessed and labeled. The whole-time load prediction model is established based on the full-time power flow data set, and the BP neural network algorithm is used for modeling and training. The model input is historical and real-time power flow data, such as line load and voltage information. The neural network model is trained through historical data, and the error feedback mechanism is used to dynamically adjust the weight parameters of the full-time load prediction model. The model output is the line load rate prediction value of the future line and the "N-1" line sensitivity prediction value. The line load rate prediction value is calculated by the prediction model, and the calculation formula is:

[0083]

[0084] Among them, m is the number of line ij, and P ij is the power flow of line ij, and P ijmax is the upper limit of the active power of line ij. The higher the load rate of the line, the greater the probability of line overload caused by power flow transfer, and the higher the overload risk coefficient of line ij. Based on the predicted value of line load, the load rate of the future state of the power network can be obtained; the ratio of the line power flow value to the maximum allowable transmission capacity can quantify the overload risk of the line; and the predicted value of the "N-1" line sensitivity is calculated based on the branch outage distribution factor (LODF), which characterizes the sensitivity of the change in the active power flow of the non-outage branch to the change in the active power flow of the outage branch when the "N-1" event occurs in the power system. For a network with N nodes and L branches, the LODF matrix is as follows:

[0085] F = B L AB -1 A T

[0086] In the formula: B L is the diagonal matrix of branch susceptance of L×L order; A is the branch-node incidence matrix of L×N order; B is the node susceptance matrix of N×N order; F is the LODF matrix of L×L order, and the element F nm in its m-th row and n-th column represents the outage distribution factor generated on line with line number n when the line with line number m has an outage fault. The calculation formula for the predicted value of the "N-1" line sensitivity is:

[0087]

[0088] In the formula: m and n are the line numbers corresponding to lines ij and pq respectively; is the change in power flow caused on line ij after line pq is disconnected, and its value can be calculated through the branch outage distribution factor; P ij0 and P pq0 are the initial power flow values of lines ij and pq before the fault; P ijmax is the maximum allowable transmission capacity of line ij; is the "N-1" sensitivity of line ij after line pq is disconnected, which characterizes the proportion of the change in power flow of the line after the fault to its power flow margin. Based on the predicted value of line load, the predicted value of the "N-1" line sensitivity of the future state of the power network can be obtained; finally, the predicted load rate and sensitivity values are compared with the real-time power flow distribution to identify the lines that may be overloaded or prone to failure in the future, and combined with dynamic analysis to generate the trend diagram of the future state power flow distribution, so as to comprehensively predict the dynamic power flow distribution of the power grid and provide high-precision prediction data support for the subsequent location selection of phase shifters.

[0089] Furthermore, predicting the future state of the current power line load includes:

[0090] Using the historical power grid power flow data as the training set and the real-time power grid power flow data as the validation set, training and generating a full-time domain load prediction model;

[0091] Analyzing the real-time power grid power flow data, obtaining the ratio of the line power flow value to the maximum allowable capacity of the line, and calculating the load margin based on the ratio of the line power flow value to the maximum allowable capacity of the line;

[0092] The full-time domain load prediction model calculates the predicted value of the line load rate based on the ratio of the line power flow value to the maximum allowable capacity of the line;

[0093] Based on the full-time domain load prediction model, establish the dynamic correlation between the full-time domain power flow change and the load margin between lines, and generate the predicted value of the line sensitivity of each line in the power network according to the dynamic correlation.

[0094] As an optimization of the above embodiment, collect the historical power grid power flow data of the power grid operation, including line load, node voltage, line topology structure, etc., use it as the training set of the full-time domain load prediction model, and at the same time use the real-time power grid power flow data as the validation set of the model. By training the neural network model, establish the mapping relationship from historical and real-time data to the dynamic power flow distribution in the future state. During the model training process, dynamically adjust the network weights and parameters; by analyzing the real-time power grid power flow data, obtain the key parameters of line operation, including the current power flow load of the line and the maximum carrying capacity of the line. On this basis, quantify the load margin of the line to reflect the remaining capacity of the line under the current operating conditions and provide support for predicting future power flow loads; use the trained full-time domain load prediction model, combined with the real-time power grid power flow data, to predict the future line load. The prediction results include the predicted value of the load rate of each line, which is used to reflect the load level of the line under future operating conditions. This process can identify the lines that may be overloaded and provide risk warnings for future power grid operation; based on the full-time domain load prediction model, further analyze the dynamic correlation between each line in the power grid, combined with the load margin and power flow distribution prediction, quantify the mutual influence degree of the power flow change between each line, establish the dynamic correlation between lines, and generate the predicted value of the line sensitivity of each line, which describes the adjustment ability and importance of the line to the power flow distribution of other lines in the power grid; provide the predicted line load rate value and line sensitivity predicted value obtained by the prediction to the subsequent steps, which can not only help identify the lines with relatively large potential risks in future operation, but also provide a scientific basis for the location selection of phase shifters. The predicted line load rate value is used to identify the lines prone to overload, and the line sensitivity predicted value is used to evaluate the key performance of the line in future power flow regulation.

[0095] Furthermore, as Figure 3 shown, it is characterized in that a power load path is established according to the dynamic power flow distribution, including:

[0096] Analyze the power flow distribution characteristics of the power grid operation based on the dynamic power flow distribution in the full time domain dimension, and identify the convergence source nodes and divergence source nodes of the power flow;

[0097] Analyze the power flow transmission relationship between the convergence source nodes and the divergence source nodes, and define the power load path construction rules based on the power flow transmission relationship with the goal of minimizing the path power flow loss and avoiding path overload;

[0098] Based on the power load path construction rules, establish the topological structure of the power load path from the divergence source nodes to the convergence source nodes;

[0099] Verify the load adaptability of the established power load path, and update the topological structure of the power load path according to the verification results.

[0100] As a preference of the above embodiments, based on the full-time-domain power grid power flow data, analyze the dynamic power flow distribution characteristics in the power grid. Through the historical and real-time power grid power flow data, combined with the predicted dynamic power flow distribution in the future state, identify the key areas of power flow in the power grid, including the converging source nodes and the diverging source nodes. The converging source nodes are the main power input points for power flow to be transmitted to external nodes, while the diverging source nodes are the main output points of power flow, representing the starting and ending points of power flow transmission. Analyze the power flow transmission relationship between the converging source nodes and the diverging source nodes, determine the main lines of power flow transmission and their influence ranges. Based on the power flow transmission relationship, define the construction rules for the power load path. The construction rules include minimizing power flow losses, that is, preferentially selecting paths with shorter transmission distances and smaller line impedances. Combine the load rate and margin information of the lines to ensure that the lines within the power load path will not be overloaded due to power flow transmission. Select lines with smaller power flow fluctuations and higher operation reliability to construct the power load path. According to the construction rules, gradually establish the topological structure of the power load path from the diverging source nodes to the converging source nodes. The path topological structure is composed of key lines and connecting nodes, ensuring that the path can meet the requirements of power flow transmission and has good operation adaptability. In the topological structure, mark the important lines within the path to provide a basis for subsequent path evaluation and phase shifter location selection. Conduct load adaptability verification on the established power load path. The verification content includes simulating the dynamic changes of power flow within the power load path to confirm whether the power load path can carry the actual power flow. Analyze the operation state of the power load path under high load or fault conditions to ensure its sufficient operation reliability. Check whether the power load path can effectively disperse the power flow to avoid local line overload. According to the results of the load adaptability verification, dynamically update and optimize the topological structure of the power load path, adjust the line priorities in the power load path, and eliminate the lines that are not suitable for power flow transmission. Introduce standby power load paths according to the verification results to improve the reliability and flexibility of the path. Mark the key nodes and lines in the optimized path topological structure to provide input data for the optimization model of phase shifter location selection.

[0101] Furthermore, evaluate the path importance of the power load path, including:

[0102] Based on the power flow distribution characteristics, perform weighted accumulation on each power load path and calculate the comprehensive power flow carrying capacity of each power load path;

[0103] Combine the power grid power flow data in the full-time-domain dimension and the comprehensive power flow carrying capacity to analyze the operation stability of the power load path and calculate the operation reliability coefficient;

[0104] Based on the topological structure of the power load path and the dynamic power flow distribution, simulate the impact of key line faults in the power load path on the overall power grid power flow change and calculate the fault sensitivity coefficient;

[0105] Generate a path importance index based on the operation reliability coefficient and the fault sensitivity coefficient, and sort the path importance indexes of each power load path to generate an important path evaluation result.

[0106] As an optimization of the above embodiment, according to the power flow distribution characteristics of the power grid, perform weighted cumulative calculation on each power load path, comprehensively consider the power flow load level and transmission capacity of the lines within the power load path, quantify the power flow carrying capacity of the path, and assign weights to the importance of each line within the power load path according to the power flow load margin. The comprehensive power flow carrying capacity of the power load path is obtained through weighted aggregation, reflecting the overall efficiency of the path in power grid power flow transmission; combining the full-time domain power grid power flow data and the comprehensive power flow carrying capacity, evaluate the stability of the path under various operating conditions, including normal operation, high load operation, and typical fault scenarios, etc. Through power flow calculation and dynamic simulation, analyze the power flow change characteristics of the path under different operating conditions, quantify the operating stability of the path, and calculate the operation reliability coefficient of the path, which is used to characterize the safety and reliability of the path during long-term operation; based on the topological structure and dynamic power flow distribution of the power load path, simulate the impact of the failure of key lines within the path on the overall power flow change of the power grid. Through power flow calculation, analyze the power flow redistribution situation caused by the failure of each key line, including the power flow change amplitude of other lines and the change amplitude of the carrying capacity of the path, so as to calculate the fault sensitivity coefficient of the path and quantify the vulnerability of the path when dealing with the failure of key lines; according to the operation reliability coefficient and the fault sensitivity coefficient of the path, generate the path importance index of each power load path. The importance index comprehensively reflects the power flow carrying capacity, operating stability, and fault resistance of the path. By sorting the importance indexes of all paths, identify the key important paths in the power grid, and obtain the link matrix based on the predicted value of the line load rate and the predicted value of the N-1 line sensitivity Its corresponding matrix element

[0107]

[0108] In the formula, γ is a balance coefficient, which is used to balance the influence of the predicted value of the "N-1" line sensitivity and the predicted value of the line load rate on the line. Normalize the matrix column by column to obtain the link matrix M. Substitute it into the following formula for iteration based on the Pagerank algorithm:

[0109] R n =βMR n-1 +(1-β)R 0 .

[0110] In the formula: R n is the PR value after the nth iteration, which is a column vector of N×1 order; R 0is the normalized form of an N×1 unit column vector. Based on the predicted line load rate and the predicted N-1 line sensitivity, the PR value sequence PR of the future state of the power network can be obtained. m1 , PR m2 ,..., PR mW . By sorting the line PR values, the path importance ranking considering "real-time state - future state" can be obtained, based on:

[0111] Y m =λ 1 PR m1 +λ 2 PR m2 +...+λ W PR mW

[0112] The comprehensive importance coefficient Y of path m is obtained. m . By considering the importance of the lines in the current and future periods of the path and sorting, according to the sorting results, an importance evaluation report of the power load path is generated, identifying the key path with the highest priority, providing an important basis for the subsequent location selection of the phase shifter.

[0113] Furthermore, as Figure 4 shown, several phase shifter location selection schemes are generated, including:

[0114] Based on the dynamic power flow distribution and phase shifter parameters, calculate the phase shifter sensitivity coefficient of each power load path, and construct a phase shifter sensitivity matrix according to the phase shifter sensitivity coefficient;

[0115] Combined with the important path evaluation results and the phase shifter sensitivity matrix, analyze the adjustment effect of the phase shifter on each power load path, and construct a set of candidate phase shifter lines;

[0116] Based on the dynamic power flow distribution in the full time domain dimension, establish a multi-objective optimization model for phase shifter location selection in combination with the phase shifter sensitivity matrix;

[0117] According to the multi-objective optimization model for phase shifter location selection, analyze the set of candidate phase shifter lines to generate several phase shifter location selection schemes.

[0118] As the preference of the above embodiment, based on the dynamic power flow distribution and phase shifter parameters, analyze the power flow adjustment effect of the phase shifter on each power load path. For each power load path, calculate the impact of installing the phase shifter on the power flow change of the target path and its surrounding paths, and obtain the phase shifter sensitivity coefficient. The phase shifter sensitivity coefficient quantitatively describes the adjustment ability of the phase shifter installed on a certain line on the power flow distribution of the whole network. Further, combine the phase shifter sensitivity coefficients of all paths to construct a phase shifter sensitivity matrix. For a network with N nodes and L branches, its phase shifter sensitivity coefficient matrix is:

[0119]

[0120] wherein is the phase shifter sensitivity coefficient on line n when installing a phase shifter on line m. Each element in the matrix represents the influence amplitude on the power flow distribution of other lines when installing a phase shifter on a certain line. Combining the important path evaluation results and the phase shifter sensitivity matrix, comprehensively analyze the adjustment effects of each power load path, and preferentially select the lines with higher phase shifter sensitivity coefficients and on important paths to construct a set of candidate phase shifter lines. The set of candidate phase shifter lines contains key lines with strong adjustment capabilities for power flow distribution, ensuring that the location selection of the phase shifter can significantly improve the balance of power flow distribution. Based on the dynamic power flow distribution in the full-time domain dimension, combined with the phase shifter sensitivity matrix, establish a multi-objective optimization model for phase shifter location selection. The optimization objectives of the multi-objective optimization model for phase shifter location selection include enhancing the power flow adjustment ability, maximizing the balance of the power grid power flow distribution, and reducing overloaded lines; improving the operation economy, minimizing the investment and operation costs of the phase shifter; enhancing the reliability of power grid operation, ensuring that the phase shifter location selection scheme still has good adjustment performance under high load or fault scenarios. The multi-objective optimization model for phase shifter location selection guarantees the feasibility of phase shifter location selection through constraint conditions, such as the physical limitations of lines, the technical parameters of phase shifters, and the operation boundaries of the power grid. Solve the set of candidate phase shifter lines through a multi-objective optimization algorithm (such as the NSGA-II algorithm or genetic algorithm). On the basis of multi-objective balance, the multi-objective optimization model for phase shifter location selection generates several phase shifter location selection schemes for different power grid operation requirements. Each phase shifter location selection scheme contains a set of phase shifter installation lines and their related operation parameters, and quantitatively evaluates the power flow adjustment ability and operation economy of the scheme.

[0121] Furthermore, calculating the phase shifter sensitivity coefficients of each power load path includes:

[0122] Based on the dynamic power flow distribution in the full-time domain dimension, obtain the basic operation data of each line in the power load path;

[0123] Based on the basic operation data and phase shifter parameters, analyze the power flow adjustment characteristics of each power load path;

[0124] According to the power flow adjustment characteristics, calculate the influence amount of the power flow change on other power load paths when installing a phase shifter on each power load path;

[0125] Perform a weighted calculation on the influence amount of the power flow change, and comprehensively consider the phase shifter parameters. Generate the phase shifter sensitivity coefficient of the corresponding power load path according to the calculation weight of the influence amount of the power flow change.

[0126] Preferably, based on the dynamic power flow distribution in the full time domain, the basic operation data of each power load path in the power grid is collected. The basic operation data includes the real-time power flow value of the line, the node voltage amplitude, the phase angle difference, the line impedance, and the capacity limit, etc., which reflects the power flow state of the path under current and historical operation conditions and provides necessary inputs for subsequent sensitivity analysis. Combining the basic operation data and the phase shifter parameters (such as the adjustment range, capacity, and angle change characteristics, etc.), the power flow adjustment characteristics of each power load path are analyzed. Specifically, the ability to adjust the power flow distribution within the path and the impact on the power flow distribution of the surrounding paths after installing a phase shifter in the path is evaluated. The analysis results will quantify the adjustment effect of the phase shifter on a specific path and provide a basis for calculating the sensitivity coefficient of the phase shifter. Based on the analysis of the power flow adjustment characteristics, when simulating the installation of a phase shifter on each power load path, the impact on the change of the power flow distribution of other paths is simulated. Specifically, after injecting different angles of adjustment by the phase shifter, the change of the power flow distribution of the target path and other paths is simulated, and the amplitude of the adjustment effect is quantified. The impact reflects the adjustment ability of the phase shifter on different paths. The impact on the power flow change of each path is weighted and calculated, taking into account the phase shifter parameters (such as capacity, sensitivity range, etc.) and path characteristics (such as line importance and transmission capacity). Through weighted calculation, the sensitivity coefficient of the phase shifter for each power load path is generated. The sensitivity coefficient of the phase shifter quantitatively characterizes the ability to adjust the power flow distribution of other paths when installing a phase shifter on the target path. The larger the sensitivity coefficient of the phase shifter, the more priority the path has in the phase shifter location selection. The generated sensitivity coefficient of the phase shifter is used as an important input data for the phase shifter location selection optimization model, providing a basis for subsequent candidate line screening and location selection scheme generation. Paths with high sensitivity coefficients of the phase shifter will be given priority to ensure that the installation location of the phase shifter can maximize the power flow adjustment of the entire network.

[0127] Furthermore, as Figure 5 shown, multi-scenario simulation verification is carried out for several phase shifter location selection schemes, including:

[0128] Based on the dynamic power flow distribution and the historical power grid power flow data, a simulation scenario covering various operation states is constructed;

[0129] According to the phase shifter parameters, the power grid topology structure, and the power flow distribution characteristics, the simulation operation parameters required for the simulation scenario are set;

[0130] For each phase shifter location selection scheme, power flow calculations are performed according to the simulation operation parameters in different simulation scenarios;

[0131] Based on the results of the power flow calculation, the power flow adjustment ability of each phase shifter location selection scheme in different simulation scenarios and the impact on the power grid operation stability are evaluated, and the location adaptability of each phase shifter location selection scheme is quantified;

[0132] Evaluate and rank the location selection schemes of each phase shifter according to the siting adaptability, and screen out the location selection schemes of phase shifters that can maintain the operating performance under multiple simulation scenarios to generate the location selection results of phase shifters.

[0133] As an optimization of the above embodiment, based on the NSGA-II algorithm, with the maximum wind power consumption rate, the maximum transmission capacity (TTC), and the investment and operating costs of controllable phase shifters as the optimization objective functions, a multi-objective optimization model for the location selection of phase shifters based on the transmission network is proposed. The NSGA-II multi-objective genetic algorithm is introduced and applied to the optimal location selection of controllable phase shifters in the transmission network:

[0134] Objective function 1: Maximum wind power consumption rate

[0135]

[0136] In the formula, c k represents the probability of scenario k, W k represents the wind power consumption under scenario k, W kmax represents the total wind power output under scenario k, S n is the number of wind power scenarios.

[0137] Objective function 2: Maximum transmission capacity

[0138] maxf 2 = TTC

[0139] Use the clustering scenario method to generate typical scenarios, take each scenario as the base state, and combine with the phase shifter sensitivity coefficient to deduce the mathematical expression of the active power of each line, and solve the TTC between the power transmission area and the power receiving area through linear programming. The specific steps are as follows:

[0140] Step1: Use the K-means clustering method to obtain n wind power scenarios, and let i = 1;

[0141] Step2: Select a typical power grid operation scenario, substitute the wind power output in scheme i into the typical operation scenario and take it as the base state, and perform a power flow calculation. Judge whether the wind power consumption rate under each scenario is 100%. If all cannot be consumed, take the magnitude of the regional sectional power flow under the maximum wind power consumption as the TTC; if all can be consumed, go to Step3;

[0142] Step3: Calculate the sensitivity coefficient S l of line l when all generators in the source area except the balancing machine change:

[0143]

[0144] In the formula, ΔE represents the change value of the output of all generators in the source area except the balancing machine;

[0145] Step 4: Combine the sensitivity coefficients of each controllable phase shifter to line l to obtain the power flow expressions for each line:

[0146]

[0147] In the formula, represents the sensitivity coefficient of the controllable phase shifter i to the active power of line l, and P l0 represents the initial active power of line l in the base-state power grid;

[0148] Step 5: Combine Step 3 and Step 4 to obtain the final power flow expressions for each line:

[0149]

[0150] Step 6: With the phase shift angles of each controllable phase shifter and ΔE as decision variables, build a linear programming model with the goal of maximizing the TTC of the regional transmission section:

[0151]

[0152] In the formula, Ω T represents the set of regional interconnection lines, and P lmax represents the thermal stability limit of line l;

[0153] Step 7: Solve the linear programming model to obtain the values of the phase shift angles and ΔE, substitute them into the base-state power grid, and then appropriately correct their values according to the continuous power flow method. Obtain the value of TTC based on the corrected power flow results of the base-state power grid, and let i = i + 1;

[0154] Step 8: If i = S n , go to Step 9; if i < S n , return to Step 2;

[0155] Step 9: After obtaining the TTC values in all scenarios, calculate the expected value of TTC, ETTC, and its expression is:

[0156]

[0157] Record the ETTC value obtained at this time as the TTC size corresponding to the phase shifter location selection scheme;

[0158] The basic principle of the continuation power flow method adopted in Step 7 is as follows: Select a certain simulation operation scenario as the basic operation state, continuously increase the output of the power generation area, and at the same time correspondingly increase the load size of the power receiving area. During the process of increasing the output of the power generation area, repeatedly calculate the power grid power flow to obtain a series of power flow solution sets, and check various constraint conditions for these power flow solution sets until the maximum output of the generator that can meet all constraint conditions is found. The magnitude of the inter-regional transmission power at this time is denoted as TTC;

[0159] Objective function 3: Investment and operation cost of controllable phase shifters

[0160] Seek the minimum sum of the investment and operation and maintenance costs of controllable phase shifters:

[0161]

[0162] C pi =γ·S pi

[0163] In the formula, C pi represents the investment cost of the controllable phase shifter i, β represents the annual operation cost coefficient, T i represents the operation time of the controllable phase shifter i, γ is the cost coefficient, and S pi is the capacity of the controllable phase shifter i.

[0164] Embodiment 2;

[0165] Based on the same inventive concept as the method for locating controllable phase shifters in a complex power grid based on dynamic power flow trend prediction in the foregoing embodiment, the present invention also provides a system for locating controllable phase shifters in a complex power grid based on dynamic power flow trend prediction. The system includes:

[0166] An analysis and prediction module, which sets the full-time domain dimension, obtains power grid power flow data and phase shifter parameters according to the full-time domain dimension, and analyzes and predicts the dynamic power flow distribution of the power grid based on the power grid power flow data;

[0167] A path evaluation module, which establishes a power load path according to the dynamic power flow distribution and evaluates the path importance of the power load path to generate an important path evaluation result;

[0168] A scheme generation module, which combines the phase shifter parameters and the important path evaluation result to construct a multi-objective optimization model for phase shifter location and generates several phase shifter location schemes;

[0169] A verification and location module, which conducts multi-scenario simulation verification on several phase shifter location schemes, verifies the applicability of the phase shifter location schemes under different power flow scenarios, and screens to obtain the phase shifter location results.

[0170] The above adjustment system in the present invention can effectively implement a method for locating controllable phase shifters in complex power grids based on dynamic power flow trend prediction. The technical effects it can achieve are as described in the above embodiments and will not be elaborated here.

[0171] Furthermore, the analysis and prediction module includes:

[0172] A time-domain definition unit, where the full time-domain dimension includes three dimensions: historical state, real-time state, and future state;

[0173] An information integration unit, which integrates power grid power flow data according to the full time-domain dimension to construct a full time-domain power grid power flow data set;

[0174] A model analysis unit, which establishes a full time-domain load prediction model. Based on the full time-domain power grid power flow data, it predicts the future state of the current power line load and generates a line load rate prediction value and a line sensitivity prediction value;

[0175] The comparison and prediction unit compares the line load rate prediction value and the line sensitivity prediction value with the power grid power flow data according to the temporal change to predict the dynamic power flow distribution.

[0176] Similarly, for the above optimization solutions of the system, they can respectively achieve the corresponding optimization effects of the method in Embodiment 1, which will not be elaborated here either.

[0177] Although the present application has been described in connection with specific features and their embodiments, it is obvious that various modifications and combinations can be made without departing from the spirit and scope of the present application. Accordingly, this specification and the drawings are merely exemplary illustrations of the present application as defined by the appended claims and are considered to cover any and all modifications, variations, combinations, or equivalents within the scope of the present application. Obviously, those skilled in the art can make various changes and modifications to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalent technologies, the present application is intended to include these changes and modifications.

Claims

1. A method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction, characterized in that: The method comprises: Setting a full time domain dimension, acquiring power grid flow data and phase shifter parameters according to the full time domain dimension, and analyzing and predicting a dynamic power flow distribution of a power grid based on the power grid flow data; Establishing a power load path according to the dynamic power flow distribution, and evaluating the path importance of the power load path to generate an important path evaluation result; Combining the phase shifter parameters and the important path evaluation results, constructing a multi-objective optimization model for phase shifter location selection, and generating several phase shifter location selection schemes; A multi-scenario simulation verification is performed on several of the phase shifter site selection schemes to verify the applicability of the phase shifter site selection schemes in different power flow scenarios, and the phase shifter site selection results are screened and obtained.

2. A method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 1, characterized in that: Analyzing and predicting the dynamic power flow distribution of the power grid based on the power grid power flow data includes: The full time domain dimension includes three dimensions: historical state, real-time state and future state; Integrate the power grid flow data according to the full time domain dimension to construct a full time domain power grid flow data set; Establishing a full-time load prediction model, based on the full-time power grid flow data set, predicting the future state of the current power line load, and generating a line load rate prediction value and a line sensitivity prediction value; The line load rate prediction value and the line sensitivity prediction value are compared with the power grid flow data according to the time state changes to predict the dynamic flow distribution.

3. A method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 2, characterized in that: Predicting the future state of the current power line load includes: The historical power grid flow data is used as a training set, and the real-time power grid flow data is used as a verification set to train and generate the full-time domain load prediction model; Analyze the real-time power grid flow data, obtain the ratio of the line flow value to the maximum allowable capacity of the line, and calculate the load margin based on the ratio of the line flow value to the maximum allowable capacity of the line; The full-time load prediction model calculates the line load rate prediction value based on the ratio of the line power flow value to the maximum allowable capacity of the line; Based on the full-time load prediction model, a dynamic correlation between the full-time power flow change between lines and the load margin is established, and the line sensitivity prediction value of each line in the power network is generated according to the dynamic correlation.

4. The method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 1, characterized in that: Establishing a power load path according to the dynamic power flow distribution includes: Analyze the power flow distribution characteristics of power grid operation based on the dynamic power flow distribution in the full time domain dimension, and identify the convergence source nodes and divergence source nodes of the power flow; Analyze the power flow transmission relationship between the convergence source node and the divergence source node, and define a power load path construction rule based on the power flow transmission relationship with the goal of minimizing path power loss and avoiding path overload; Based on the power load path construction rule, establishing a topological structure of the power load path from the divergent source node to the convergent source node; The load adaptability of the established power load path is verified, and the topology of the power load path is updated according to the verification result.

5. The method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 4, characterized in that: An assessment of the path importance of the power load path is performed, including: Based on the power flow distribution characteristics, weighted accumulation is performed on each of the power load paths, and a comprehensive power flow carrying capacity of each of the power load paths is calculated; Combining the power grid flow data of the full time domain dimension and the comprehensive flow carrying capacity, analyzing the operation stability of the power load path, and calculating the operation reliability coefficient; Based on the topological structure of the power load path and the dynamic power flow distribution, simulating the impact of key line failures in the power load path on the overall power flow change of the power grid, and calculating the fault sensitivity coefficient; A path importance index is generated according to the operation reliability coefficient and the fault sensitivity coefficient, and the path importance index of each of the power load paths is sorted to generate the important path evaluation result.

6. The method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 1, characterized in that: Generate several phase shifter siting options, including: Based on the dynamic power flow distribution and the phase shifter parameters, calculating the phase shifter sensitivity coefficient of each of the power load paths, and constructing a phase shifter sensitivity matrix according to the phase shifter sensitivity coefficients; Combining the important path evaluation result and the phase shifter sensitivity matrix, analyzing the regulation effect of the phase shifter on each of the power load paths, and constructing a phase shifter candidate line set; Based on the dynamic power flow distribution in the full time domain dimension and in combination with the phase shifter sensitivity matrix, a multi-objective optimization model for phase shifter location is established; The phase shifter candidate line set is analyzed according to the phase shifter site selection multi-objective optimization model to generate a plurality of phase shifter site selection schemes.

7. A method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 6, characterized in that: Calculating the phase shifter sensitivity coefficient of each of the power load paths, including: Based on the dynamic power flow distribution in the full time domain dimension, basic operation data of each line in the power load path is obtained; Analyzing the power flow regulation characteristics of each of the power load paths based on the basic operation data and the phase shifter parameters; According to the power flow regulation characteristics, calculating the influence of each power load path on the power flow change of other power load paths when a phase shifter is installed; The power flow change influence amount is weightedly calculated, the phase shifter parameters are integrated, and the power load path is weighted according to the calculation weight of the power flow change influence amount to generate the phase shifter sensitivity coefficient corresponding to the power load path.

8. The method for selecting a site for a controllable phase shifter in a complex power grid based on dynamic power flow trend prediction according to claim 1, characterized in that: A multi-scenario simulation verification is performed on several of the phase shifter site selection schemes, including: Based on the dynamic power flow distribution and the historical power grid power flow data, construct simulation scenarios covering various operating states; According to the phase shifter parameters, the grid topology and the power flow distribution characteristics, the simulation operation parameters required by the simulation scenario are set; Performing power flow calculation for each of the phase shifter site selection schemes in different simulation scenarios according to the simulation operation parameters; According to the results of the power flow calculation, the power flow regulation capability and the impact on the power grid operation stability of each of the phase shifter site selection schemes under different simulation scenarios are evaluated, and the site selection adaptability of each of the phase shifter site selection schemes is quantified; According to the site selection adaptability, the phase shifter site selection schemes are evaluated and ranked, and the phase shifter site selection schemes that can maintain the operating performance under the multiple simulation scenarios are screened out to generate the phase shifter site selection results.

9. A complex power grid controllable phase shifter site selection system based on dynamic power flow trend prediction, characterized in that: The system comprises: The analysis and prediction module sets the full time domain dimension, obtains the power grid flow data and phase shifter parameters according to the full time domain dimension, and analyzes and predicts the dynamic power flow distribution of the power grid based on the power grid flow data; The path assessment module establishes the power load path according to the dynamic power flow distribution, evaluates the path importance of the power load path, and generates the important path assessment result; The scheme generation module combines the phase shifter parameters and the important path evaluation results to build a multi-objective optimization model for phase shifter location selection and generate several phase shifter location selection schemes; Verify the site selection module, conduct multi-scenario simulation verification on several phase shifter site selection schemes, verify the applicability of the phase shifter site selection schemes under different power flow scenarios, and screen out the phase shifter site selection results.

10. A complex power grid controllable phase shifter site selection system based on dynamic power flow trend prediction according to claim 9, characterized in that: The analysis and prediction module includes: Time domain definition unit, the full time domain dimension includes historical state, real-time state and future state; The information integration unit integrates the power grid flow data according to the full time domain dimension and constructs a full time domain power grid flow data set; The model analysis unit establishes a full-time load prediction model, predicts the future state of the current power line load based on the full-time power grid flow data, and generates a line load rate prediction value and a line sensitivity prediction value; The comparison prediction unit compares the line load rate prediction value and the line sensitivity prediction value with the power grid flow data according to the time state change, and predicts the dynamic flow distribution.

Images