A method and device for evaluating a power grid planning scheme

Abstract

This invention discloses a method and apparatus for evaluating power grid planning schemes, belonging to the field of power system operation and planning. The method includes: acquiring basic data of a benchmark power grid and the power grid planning scheme to be evaluated; constructing a unit combination model based on the cost parameters of each unit and the length of the time window, with the goal of minimizing the total start-up and shutdown costs and the total operating costs, and solving the model under a first constraint; constructing an economic dispatch model based on the total number of units, the length of the time window, and the fitting coefficient of the output cost of each unit, with the goal of minimizing the total operating costs of the power grid planning scheme to be evaluated, and updating the first constraint to generate a second constraint; solving the economic dispatch model under the second constraint; and determining the evaluation result of the power grid planning scheme to be evaluated based on the solution result. Therefore, by implementing this invention, the problem of low accuracy in the evaluation results of power grid planning schemes in the prior art can be solved.

Description

Technical Field

[0001] This invention relates to the field of power system operation and planning, and in particular to a method and apparatus for evaluating power grid planning schemes. Background Technology

[0002] With the continuous growth of electricity load and the large-scale integration of new energy sources, the power grid structure is becoming increasingly complex. New power grid planning schemes need to be developed frequently to improve the power grid architecture and ensure reliable power supply. Before the power grid planning scheme is implemented, it is necessary to conduct a reasonable evaluation of the scheme.

[0003] Currently, existing evaluation methods for power grid planning schemes typically only verify the rationality of the power grid topology without conducting a systematic evaluation in conjunction with the load characteristics of the power grid. They also fail to fully consider factors such as unit operation constraints, start-up and shutdown costs, and operating costs, thus failing to accurately reflect the actual operating effect of the power grid planning scheme and resulting in low accuracy in the evaluation results. Summary of the Invention

[0004] This invention provides a method and apparatus for evaluating power grid planning schemes, which can solve the problem of low accuracy in the evaluation results of power grid planning schemes in the prior art.

[0005] The present invention provides a method for evaluating power grid planning schemes, comprising: Obtain historical source-load data from the benchmark power grid; Obtain the total number of generating units, the length of the time window, the operating constraints of each generating unit, the cost parameters of each generating unit, the grid topology parameters, and the theoretical power generation of each generating unit in each time period for the power grid planning scheme to be evaluated; Based on historical source load data, source load sample data for each time period within the time window are obtained; Based on the cost parameters of each generating unit and the length of the time window, a generating unit combination model is constructed with the goal of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and based on the operating constraint parameters of each generating unit, the power grid topology parameters, the source-load sample data, and the length of the time window, the first constraint condition is constructed. Under the first constraint, the unit combination model is solved to obtain the start-up and shutdown status of each unit in each time period, the maximum output limit and the minimum output limit of each unit in each time period. Based on the total number of generating units, the length of the time window, and the preset fitting coefficients for the output cost of each generating unit, an economic dispatch model is constructed with the goal of minimizing the total operating cost of the power grid planning scheme to be evaluated within the time window. Based on the start-up and shutdown status, the maximum and minimum output limits of each generating unit in each time period, the first constraint is updated to generate the second constraint. Under the second constraint, the economic dispatch model is solved to obtain the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. Based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period, the evaluation results of the power grid planning scheme to be evaluated are determined.

[0006] Furthermore, historical source-load data includes: historical renewable energy output data and historical load data; Based on historical source-load data, source-load sample data for each time period within the time window are obtained, including: Feature extraction was performed on historical renewable energy output data to obtain historical output characteristics including daily maximum output, daily minimum output, daily average output volatility, and daily peak-to-valley difference. Feature extraction is performed on historical load data to obtain historical load characteristics including daily maximum load, daily minimum load, daily average load fluctuation rate, and daily load peak-to-valley difference; Based on historical power output characteristics and historical load characteristics, historical renewable energy power output data and historical load data are classified into several typical scenarios. For each typical scenario, several source-load sample data corresponding to the typical scenario are generated to obtain the source-load sample data for each time period in the time window.

[0007] Furthermore, the cost parameters for each unit include: start-up and shutdown costs, operating cost coefficients, and fixed operating costs for each unit; The unit combination model includes: in, The total number of units. For the unit's index, For time period index, The length of the time window, For the unit Start-up and shutdown costs For the unit During the period Start-stop status, For the unit During the period Start-stop status, Let g be the operating cost coefficient of unit g. For the unit During the period The predictions were helpful and productive. unit Fixed operating costs.

[0008] Furthermore, the economic scheduling model includes: in, The total number of units. For the unit's index, For time period index, The length of the time window, , , The fitting coefficients for the output cost of each unit. For the unit During the period The predictions were helpful and productive.

[0009] Furthermore, based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period, the evaluation results of the power grid planning scheme to be evaluated are determined, including: Based on the marginal electricity price time series, the spatiotemporal fluctuation index of electricity price is determined; based on the power flow of each line in each time period, the spatial concentration index of congestion cost is determined; based on the predicted active power output of each unit in each time period and the theoretical power generation of new energy, the spatiotemporal consumption index of new energy is determined. The evaluation results of the power grid planning scheme to be evaluated are determined based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy sources.

[0010] Furthermore, the determination of the spatiotemporal fluctuation index of electricity prices based on the marginal electricity price time series includes: For each node, the marginal electricity price time series corresponding to the node is divided into several price levels, and the probability of each price level occurring is calculated. The temporal entropy of each node is obtained based on the probability of each electricity price level occurring at each node. Based on the time series of marginal electricity prices, the electricity price of each node in each time period is determined; Based on the electricity price of each node in each time period, the average electricity price of all nodes in a single time period and the degree of price inequality among all nodes in a single time period are calculated. For each time period, the Gini coefficient for electricity price is determined based on the average electricity price and the degree of electricity price inequality for that time period. The temporal entropy of each node and the Gini coefficient of electricity price for each time period are weighted and fused to determine the spatiotemporal fluctuation index of electricity price for the power grid planning scheme to be evaluated.

[0011] Furthermore, based on the power flow of each line in each time period, the spatial concentration index of congestion costs is determined, including: The marginal electricity price time series of each node is decomposed to extract the congestion price of each line; Based on the predicted active power output of each unit in each time period, the congestion price of each line, and the power flow of each line in each time period, the total line congestion cost of each line in the power grid planning scheme to be evaluated within the time window is obtained. Obtain the power transfer distribution factor between each generating unit and its corresponding lines from the power grid topology parameters; Based on the total line congestion cost of each line within the time window, a congestion cost calculation operation is performed for each unit to obtain the congestion cost of each unit for each line. For each generating unit, the congestion costs of all lines corresponding to the generating unit are summed to obtain the first total congestion cost of each generating unit for all corresponding lines in the power grid planning scheme to be evaluated. Obtain the second total congestion cost for each unit across all corresponding lines in the baseline grid; For each unit, the change in total congestion cost for each unit is obtained based on the second total congestion cost and the first total congestion cost; Based on the change in the total congestion cost of each unit, the spatial concentration index of congestion cost of the power grid planning scheme to be evaluated is obtained. The blocking cost calculation operation includes: Use the current unit as the target unit; The lines corresponding to the target generator set are designated as the target lines. The target generator units and each target line are respectively paired to form several generator-line pairs; For each generator line pair, the congestion cost of the generator line pair is determined based on the total line congestion cost corresponding to the target line in the generator line pair, the power transfer distribution factor corresponding to the generator line pair, and the predicted active power output of the target generator in the generator line pair within the time window, thus obtaining the current unit's congestion cost for the target line.

[0012] Furthermore, based on the predicted active power output of each unit in each time period and the theoretical power generation of new energy sources, the spatiotemporal consumption index of new energy sources is determined, including: The predicted active power output of each unit in each time period is summed to obtain the total new energy output of all units within the time window; The theoretical renewable energy power generation of each unit at each time period is summed to obtain the total theoretical renewable energy output of all units within the time window. The amount of abandoned renewable energy is obtained based on the total output of renewable energy and the theoretical total output of renewable energy. Based on the total output of new energy sources and the amount of new energy curtailment, the initial new energy consumption rate of all units within the time window is obtained; For each time period, the dynamic current carrying capacity of each line in each time period is obtained based on the preset maximum allowable temperature of the line, the standard ambient temperature, the preset meteorological sample data, the preset wind speed correction coefficient, and the static critical value of current carrying capacity under standard conditions. For each line, in each time period corresponding to the current line, the dynamic thermal setpoint of the current line in the current time period is obtained based on the preset three-phase power conversion factor, the dynamic current carrying capacity of the line in the current time period, and the preset rated voltage of the line; the dynamic thermal setpoint is normalized to obtain the normalized thermal setpoint of each line in each time period. The source-load sample data and the preset meteorological sample data are input into the hidden blockage prediction model so that the hidden blockage prediction model can output the instantaneous blockage frequency of each line in each time period. For each time period in each line, the time period in which the instantaneous blocking frequency is greater than the normalized thermal setpoint is selected as the first candidate time period; For each line, the first candidate time periods that are consecutive are connected to obtain several second candidate time periods; the second candidate time periods that are less than the first time threshold are regarded as hidden blocking time periods. The total number of concealed blocking periods across all lines is taken as the concealment count; the duration of each concealed blocking period is taken as the duration of each concealed blocking event. For each hidden blockage period, the reduction in new energy output for each hidden blockage period is determined based on the theoretical power generation of new energy sources corresponding to the hidden blockage period and the predicted active power output of each unit. The correction factor for the initial renewable energy absorption rate is determined based on the number of concealed events, the duration of each concealed blockage event, and the amount of renewable energy output reduction during each concealed blockage period. Based on the correction factor and the initial renewable energy absorption rate, the corrected renewable energy absorption rate is determined. The renewable energy absorption rate is used as the renewable energy spatiotemporal absorption index for the power grid planning scheme to be evaluated.

[0013] Furthermore, based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy sources, the evaluation results of the power grid planning scheme to be evaluated are determined, including: A comprehensive evaluation vector is generated based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal consumption index of new energy sources. Based on the comprehensive evaluation vector and the preset benchmark scheme index vector, the comprehensive deviation degree is obtained; If the overall deviation is less than the preset deviation threshold, the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme is approved. If the overall deviation is greater than or equal to the preset deviation threshold, the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme fails.

[0014] Another embodiment of the present invention provides an evaluation device for power grid planning schemes, comprising: a historical source-load data determination module, a parameter determination module, a source-load sample data determination module, a unit combination model determination module, a first solution module, an economic dispatch model determination module, a second solution module, an evaluation parameter determination module, and an evaluation module; The historical source load data determination module is used to obtain historical source load data of the reference power grid; The parameter determination module is used to obtain the total number of generating units, the length of the time window, the operating constraint parameters of each generating unit, the cost parameters of each generating unit, the grid topology parameters, and the theoretical power generation of each generating unit in each time period for the power grid planning scheme to be evaluated. The source load sample data determination module is used to obtain source load sample data for each time period in the time window based on historical source load data. The unit combination model determination module is used to construct a unit combination model based on the cost parameters of each unit and the length of the time window, with the goal of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and to construct the first constraint condition based on the operating constraint parameters of each unit, the power grid topology parameters, the source-load sample data, and the length of the time window. The first solution module is used to solve the unit combination model under the first constraint conditions to obtain the start-up and shutdown status of each unit in each time period, the maximum output limit and the minimum output limit of each unit in each time period. The economic dispatch model determination module is used to construct an economic dispatch model based on the total number of generating units, the length of the time window, and the preset fitting coefficients of the output cost of each generating unit, with the goal of minimizing the total operating cost of the power grid planning scheme to be evaluated within the time window. It also updates the first constraint based on the start-up and shutdown status, the maximum output limit and the minimum output limit of each generating unit in each time period, and generates the second constraint condition. The second solution module is used to solve the economic dispatch model under the second constraint to obtain the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. The evaluation module is used to determine the evaluation results of the power grid planning scheme to be evaluated based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period.

[0015] The following benefits can be obtained by implementing the present invention: This invention provides a method and apparatus for evaluating power grid planning schemes. The method includes: acquiring historical source-load data of a benchmark power grid; acquiring the total number of generating units, the length of the time window, the operating constraints of each generating unit, the cost parameters of each generating unit, the power grid topology parameters, and the theoretical renewable energy generation of each generating unit in each time period for the power grid planning scheme to be evaluated; obtaining source-load sample data for each time period within the time window based on the historical source-load data; constructing a generating unit combination model based on the cost parameters of each generating unit and the length of the time window, with the objective of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and constructing a first constraint condition based on the operating constraints of each generating unit, the power grid topology parameters, the source-load sample data, and the length of the time window; and solving the generating unit combination model under the first constraint condition to obtain the theoretical renewable energy generation of each generating unit in each time period. The evaluation process involves considering the start-up and shutdown status of each time period, the maximum and minimum output limits of each generating unit during each time period, the total number of generating units, the length of the time window, and the preset fitting coefficients for the output costs of each generating unit. An economic dispatch model is constructed with the goal of minimizing the total operating cost of the power grid planning scheme under evaluation within the time window. Based on the start-up and shutdown status, and the maximum and minimum output limits of each generating unit during each time period, the first constraint is updated to generate the second constraint. Under the second constraint, the economic dispatch model is solved to obtain the marginal electricity price time series of each node, the predicted active power output of each generating unit during each time period, and the power flow of each line during each time period. Based on the marginal electricity price time series of each node, the predicted active power output of each generating unit during each time period, and the power flow of each line during each time period, the evaluation results of the power grid planning scheme under evaluation are determined.Compared with existing technologies, this invention no longer simply verifies the rationality of the power grid topology of a power grid planning scheme. Instead, based on the cost parameters of each generating unit and the length of the time window, it constructs a generating unit combination model with the objective of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme within the time window. Then, based on the operating constraint parameters of each generating unit, the power grid topology parameters, source-load sample data, and the length of the time window, it constructs the first constraint condition and solves the generating unit combination model to predict the start-up and shutdown states of each generating unit in each time period. Next, based on the total number of generating units, the length of the time window, and the preset fitting coefficients for the output costs of each generating unit, it constructs an economic dispatch model with the objective of minimizing the total operating costs of the power grid planning scheme within the time window. The constraints are updated based on the start-up and shutdown states predicted by the generating unit combination model, resulting in the second constraint condition of the economic dispatch model. The economic dispatch model is then solved, yielding the marginal electricity price time series of each node, the predicted active power output of each generating unit in each time period, and the power flow of each line in each time period. Finally, based on the theoretical power generation of new energy sources... This invention evaluates power grid planning schemes by using the marginal electricity price time series output by the economic dispatch model, the power flow of each line in each time period, and the predicted active power output of each unit in each time period. Because this invention incorporates the cost parameters and time window length of each unit when constructing the unit combination model, and simultaneously constructs a comprehensive first constraint based on the operating constraint parameters of each unit and the power grid topology parameters, it ensures the accuracy of the unit start-up and shutdown state solution. Furthermore, when constructing the economic dispatch model, this application updates the first constraint based on the start-up and shutdown state output by the unit combination model to obtain the second constraint. This allows for locking the range of units that can participate in dispatch based on the accurate start-up and shutdown state of each unit, ensuring that the constraint conditions completely match the accurate operating state of the units. This ensures the relevance and accuracy of the economic dispatch solution, ultimately accurately solving for the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. Using this as the evaluation basis, it completely overcomes the shortcomings of existing technologies that only simply verify the topology, thus effectively improving the accuracy of the power grid planning scheme evaluation results. Attached Figure Description

[0016] To more clearly illustrate the technical solution of this application, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0017] Figure 1 This is a flowchart illustrating an evaluation method for a power grid planning scheme provided in an embodiment of the present invention; Figure 2 This is a schematic diagram of the structure of an evaluation device for power grid planning schemes provided in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains; the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the application; the terms “comprising” and “having”, and any variations thereof, in the specification, claims, and foregoing description of the drawings are intended to cover non-exclusive inclusion.

[0020] In the description of the embodiments of this application, technical terms such as "first" and "second" are used only to distinguish different objects and should not be construed as indicating or implying relative importance or implicitly specifying the number, specific order, or primary and secondary relationship of the indicated technical features. In the description of the embodiments of this application, "multiple" means two or more, unless otherwise explicitly defined.

[0021] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this application. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0022] In the description of the embodiments in this application, the term "and / or" is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, and B existing alone. Additionally, the character " / " in this document generally indicates that the preceding and following related objects have an "or" relationship.

[0023] In the description of the embodiments of this application, the term "multiple" refers to two or more (including two), similarly, "multiple sets" refers to two or more (including two sets), and "multiple pieces" refers to two or more (including two pieces).

[0024] In the description of the embodiments of this application, unless otherwise expressly specified and limited, technical terms such as "installation," "connection," "joining," and "fixing" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. For those skilled in the art, the specific meaning of the above terms in the embodiments of this application can be understood according to the specific circumstances.

[0025] See Figure 1 To address the issue of low accuracy in evaluating power grid planning schemes in existing technologies, an embodiment of the present invention provides a method for evaluating power grid planning schemes, comprising: S101: Obtain historical source-load data of the reference power grid.

[0026] The historical source-load data includes: historical renewable energy output data and historical load data.

[0027] Specifically, historical renewable energy output data refers to the time-series data of actual power generation output of renewable energy plants such as wind power and photovoltaic power in the benchmark grid for each time period, while historical load data refers to the time-series data of power consumption of each load node in the benchmark grid for each time period.

[0028] In one possible implementation, historical renewable energy output data and historical load data are selected from relevant data of the benchmark power grid for at least three consecutive years. Renewable energy output data is obtained from the SCADA system of each renewable energy power station at a sampling frequency of once per hour, and historical load data is obtained from the power grid dispatch automation system at a sampling frequency of once per hour. It should be noted that the SCADA system is a dedicated automation system used by renewable energy power stations, substations, and power grids to collect equipment operation data in real time and remotely monitor and control equipment.

[0029] S102: Obtain the total number of generating units, the length of the time window, the operating constraint parameters of each generating unit, the cost parameters of each generating unit, the grid topology parameters, and the theoretical power generation of each generating unit in each time period for the power grid planning scheme to be evaluated. Furthermore, the cost parameters for each unit include: start-up and shutdown costs, operating cost coefficients, and fixed operating costs for each unit.

[0030] Specifically, the total number of generating units refers to the total number of all wind power and photovoltaic new energy generating units within the power grid planning scheme to be evaluated. All generating units mentioned in this application are new energy generating units. The length of the time window is the total duration of the continuous time series selected for power grid evaluation and analysis (i.e., the simulation evaluation duration corresponding to the power grid planning scheme to be evaluated). The operating constraint parameters of each generating unit are the operating boundary parameters such as the upper and lower limits of output and the ramp rate that the new energy generating units must follow during operation. The cost parameters of each generating unit are the economic parameters characterizing the power generation and maintenance costs of the new energy generating units. The power grid topology parameters are the basic parameters describing the layout of power grid nodes, the connection relationship of lines, and the electrical attributes of lines. The theoretical power generation of each generating unit in each time period is the maximum theoretical output that each new energy generating unit can generate, which is not constrained by power grid flow and is determined only by natural resource conditions. The start-up and shutdown cost is the one-time cost incurred during the commissioning and shutdown of the new energy generating units. The operating cost coefficient is the corresponding coefficient for calculating the power generation and maintenance expenses of the new energy generating units. The fixed operating cost is the inherent basic cost incurred by the new energy generating units in maintaining their grid-connected standby state.

[0031] S103: Based on historical source load data, obtain source load sample data for each time period within the time window; Furthermore, historical source-load data includes: historical renewable energy output data and historical load data; Furthermore, based on historical source-load data, source-load sample data for each time period within the time window are obtained, including: Feature extraction was performed on historical renewable energy output data to obtain historical output characteristics including daily maximum output, daily minimum output, daily average output volatility, and daily peak-to-valley difference. Feature extraction is performed on historical load data to obtain historical load characteristics including daily maximum load, daily minimum load, daily average load fluctuation rate, and daily load peak-to-valley difference; Based on historical power output characteristics and historical load characteristics, historical renewable energy power output data and historical load data are classified into several typical scenarios. For each typical scenario, several source-load sample data corresponding to the typical scenario are generated to obtain the source-load sample data for each time period in the time window.

[0032] In one possible implementation, feature extraction of historical load data includes: First, organizing daily renewable energy output curves based on historical renewable energy output data, and then organizing daily renewable energy output curves and daily load curves respectively; extracting daily maximum value, daily minimum value, daily average volatility and daily peak-to-valley difference from the two daily time-series curves to obtain historical output characteristics and historical load characteristics.

[0033] In one possible implementation, historical renewable energy output data and historical load data are classified into several typical scenarios using the K-means clustering algorithm. Specifically, historical output features and historical load features are integrated into a feature vector and then input into the K-means clustering algorithm. The elbow rule is used to determine the optimal number of clusters, dividing the daily renewable energy output curve and load curve into three typical scenario curve categories: normal scenario (60%), anti-peak shaving scenario (30%), and extreme weather scenario (10%). Among these, the anti-peak shaving scenario... The scenario refers to a situation where the output of new energy sources and the load demand show an inverse trend, which usually occurs in spring and autumn. The extreme weather scenario refers to a situation where continuous rainy weather, windless weather, or extreme high temperature or low temperature cause the output of new energy sources to deviate significantly from the normal value. Then, within the curve range of each scenario, the Markov Chain Monte Carlo (MCMC) method is used to construct a state transition probability matrix based on the historical curve samples within the scenario. After Monte Carlo sampling, multiple source-load time series curve samples that conform to the scenario are generated, and finally, the source-load sample data of new energy output and load time series corresponding to each time period of the time window are formed.

[0034] S104: Based on the cost parameters of each unit and the length of the time window, construct a unit combination model with the goal of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and construct the first constraint condition based on the operating constraint parameters of each unit, the power grid topology parameters, the source-load sample data, and the length of the time window. Furthermore, the cost parameters for each unit include: start-up and shutdown costs, operating cost coefficients, and fixed operating costs for each unit; The unit combination model includes: (1) in, The total number of units. For the unit's index, For time period index, The length of the time window, For the unit Start-up and shutdown costs For the unit During the period Start-stop status, For the unit During the period Start-stop status, Let g be the operating cost coefficient of unit g. For the unit During the period The predictions were helpful and productive. unit Fixed operating costs.

[0035] It should be noted that in the unit combination model (also known as the safety-constrained unit combination model, SCUC model), the total number of units, the length of the time window, and the units... Start-up and shutdown costs (i.e., start-up and shutdown costs of each unit), operating cost coefficient of unit g, unit The fixed operating cost is the known fixed parameters obtained in step S102, while the unit During the period The start / stop status (indicating whether the unit is in a stopped or running state) of the generator set. During the period Start-up and shutdown status of the unit During the period The predicted active power output is an unknown quantity to be solved.

[0036] In one possible implementation, the operating constraints of each generating unit include: the minimum output of each generating unit, the maximum output of each generating unit, the ramp rate of each generating unit, the minimum start-up time of each generating unit, the minimum downtime of each generating unit, and the power flow limit of each line; the power grid topology parameters include: the power transfer distribution factor of each generating unit to each line, the power transfer distribution factor of each load node to each line, and the total number of load nodes. In one possible implementation, the first constraint includes the following: The power balance constraint is: (2) in, For the unit During the period The predictions were helpful and productive. Let t be the total system load. The total number of units. This is the index for the generator set.

[0037] The upper and lower limits of the unit output are constrained as follows: (3) in, For the unit Minimum output, For the unit Maximum output For the unit During the period Start-stop status, For the unit During the period The predictions were helpful and productive.

[0038] The unit's ramp-up constraint is: (4) in, For the unit The rate of ascent, The length of each time period t (i.e., the time interval between two evaluation times). For the unit During the period The predictions were helpful and productive. For the unit During the period The predictions were helpful and effective; The minimum start-stop time constraint is: , (5) in, For the unit During the period Start-stop status, For the unit During the period Start-stop status, For the unit Minimum boot time, For the unit Minimum downtime; For the unit The device has been continuously powered on for a certain period of time. For the unit The duration of the downtime, specifically... This refers to the duration during which the generator unit has been continuously powered on before entering the current time period t. It is the duration during which the unit has been continuously shut down before entering the current time period t.

[0039] The power flow constraints of the line are: (6) in, The total number of units. For the unit's index, For the unit For the line The power transfer distribution factor, For the unit During the period The predictions were helpful and productive. For load nodes, This represents the total number of load nodes. For load nodes For the line The power transfer distribution factor, For load nodes During the period Active load, For time period index, For the line The upper limit of the trend.

[0040] It should be noted that in the first constraint, the total number of generating units, the total number of load nodes, the total system load of the time period, the active load of the load nodes in the time period, the minimum output of the generating units, the maximum output of the generating units, the ramp rate of the generating units, the length of each time period t, the minimum start-up time of the generating units, the minimum downtime of the generating units, the continuous start-up time of the generating units, the continuous downtime of the generating units, the power transfer distribution factor of the generating units to the line, the power transfer distribution factor of the load nodes to the line, and the upper limit of the power flow of the line are known quantities.

[0041] Specifically, the total number of load nodes, the power transfer distribution factor of generating units to lines, and the power transfer distribution factor of load nodes to lines are the power grid topology parameters of the power grid planning scheme to be evaluated; the total system load for a given time period is the system-level total load time series value obtained by statistically summarizing the source-load sample data generated by scenario clustering and MCMC sampling based on historical source-load data, and the source-load sample data has been generated and determined before the optimization model is solved; the active power load of load nodes in a given time period is the node-level load time series value extracted by directly matching load nodes and time periods based on the source-load sample data generated by scenario clustering and MCMC sampling based on historical source-load data; the minimum output of each generating unit, the maximum output of each generating unit, the ramp rate of each generating unit, the minimum start-up time of each generating unit, and the other parameters are also considered. The minimum downtime of the generating units and the upper limit of power flow for each line are operating constraint parameters for each generating unit. The length of each time period t is an inherent parameter of the power grid specification scheme determined by the time window length and time period division through uniform division. The continuous operating time and continuous downtime of the generating units are historical state parameters calculated recursively based on the start-up and shutdown states of the preceding time periods. They reflect the continuous operating or shutdown duration that the generating units have maintained before entering the current time period. Since this model adopts a rolling solution method in chronological order, the start-up and shutdown states of all preceding time periods have been determined when solving the start-up and shutdown states and output decisions of the current time period. Therefore, the corresponding continuous operating time and continuous downtime have become fixed values ​​and do not need to be solved through optimization. Hence, they are known quantities.

[0042] It should be noted that in the first constraint, , , , The unknown quantity to be solved.

[0043] S105: Under the first constraint, solve the unit combination model to obtain the start-up and shutdown status of each unit in each time period, the maximum output limit and the minimum output limit of each unit in each time period.

[0044] Specifically, the unit combination model is comprehensively optimized and iteratively solved within the feasible region defined by the first constraint. During the solution process, the optimal start-stop state of each unit in each time period is first optimized and determined. While maintaining the determined start-stop states, the feasible output range of each unit in each time period is then tightened and redefined in conjunction with the system reserve capacity requirement. The system reserve capacity requirement is used to smooth out load forecast deviations, random fluctuations in renewable energy output, and to cope with uncertain disturbances to the power grid such as sudden equipment failures, meeting the power regulation margin required for grid operation. The final maximum and minimum output limits of each unit in each time period are the actual operating output upper and lower limits formed after reserving system reserve capacity based on the unit's rated output limit. They are not the inherent rated output extreme values ​​of the units, but only represent the effective adjustment range in which the units can participate in the subsequent refined output allocation of economic dispatch under the current given start-stop state. This provides a fixed start-stop benchmark and clear, controllable unit output constraint boundaries for the subsequent economic dispatch model solution.

[0045] S106: Based on the total number of generating units, the length of the time window, and the preset fitting coefficients for the output cost of each generating unit, an economic dispatch model is constructed with the goal of minimizing the total operating cost of the power grid planning scheme to be evaluated within the time window. The first constraint is updated based on the start-up and shutdown status, the maximum output limit and the minimum output limit of each generating unit in each time period, and the second constraint is generated. Furthermore, the economic scheduling model includes: (7) in, The total number of units. For the unit's index, For time period index, The length of the time window, , , The fitting coefficients for the output cost of each unit. For the unit During the period The predictions were helpful and productive.

[0046] In one possible implementation, the first constraint is updated based on the start / stop state to generate a second constraint, including: Based on the start-up and shutdown status, the upper and lower limits of unit output in the first constraint are updated to generate the second constraint; Specifically, regarding the upper and lower limits of unit output constraints (i.e., formula (3)), during the solution process of the unit combination model, It is an unknown quantity; however, in the process of solving the economic dispatch model, since the start-up and shutdown states are solved in the unit combination model, therefore... In the process of solving the economic dispatch model, it is used as a known quantity in the solution. Furthermore, for the upper and lower limits of unit output constraints, the inherent rated maximum output of the units originally configured in the economic dispatch model is discarded. With the rated minimum output limit The maximum and minimum output limits of each unit in each time period, which are redefined after considering the system's reserve capacity requirements, are directly used as the new upper and lower limit constraints of unit output in the economic dispatch model. This limits the adjustable operating range of each unit under the given start-stop conditions. The other constraints, such as power balance constraints, unit ramping constraints, minimum start-stop time constraints, and line power flow constraints, remain unchanged in their original formulas. Only the upper and lower limit parameters of output are replaced to update the constraints. Then, the optimization solution of the economic dispatch model is carried out within the tightened feasible region.

[0047] S107: Under the second constraint, the economic dispatch model is solved to obtain the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. Specifically, within the feasible domain defined by the updated and improved second constraint, the economic dispatch model is optimized and iteratively solved with the goal of optimizing the power generation operating cost. The predicted active power output of each unit that can stably connect to the grid in each time period is calculated. Then, combined with the load supply and demand balance relationship of the entire grid and the nodal electricity price accounting mechanism, the marginal electricity price time series corresponding to each node is generated. At the same time, the power flow distribution results of each transmission line in each time period are obtained by coupling the solution based on the power flow distribution principle of the power grid. This can provide complete basic data support for power grid operation benefit assessment, electricity price analysis and line safety operation verification.

[0048] In one possible implementation, during the solution of the economic scheduling model, during periods of high incidence of covert tracking events, the line power flow constraints in the economic scheduling model (i.e., formula (6)) are... Using dynamic thermal setpoints instead of static thermal setpoints, that is... The fixed static thermal setpoint will no longer be used. Instead, it is replaced by a dynamic thermal settling value that changes dynamically with ambient temperature and wind speed. This dynamic thermal setpoint The formula is: (8) in, This is a static thermal setpoint. This is the maximum allowable temperature for the circuit (typically 90°C). For a moment Ambient temperature; Standard ambient temperature (typically 40°C); For a moment wind speed; This is the wind speed correction factor (usually taken as 0.1 to 0.3).

[0049] S108: Based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period, determine the evaluation result of the power grid planning scheme to be evaluated. Furthermore, based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period, the evaluation results of the power grid planning scheme to be evaluated are determined, including: Based on the marginal electricity price time series, the spatiotemporal fluctuation index of electricity price is determined; based on the power flow of each line in each time period, the spatial concentration index of congestion cost is determined; based on the predicted active power output of each unit in each time period and the theoretical power generation of new energy, the spatiotemporal consumption index of new energy is determined. The evaluation results of the power grid planning scheme to be evaluated are determined based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy sources.

[0050] Furthermore, the determination of the spatiotemporal fluctuation index of electricity prices based on the marginal electricity price time series includes: For each node, the marginal electricity price time series corresponding to the node is divided into several price levels, and the probability of each price level occurring is calculated. The temporal entropy of each node is obtained based on the probability of each electricity price level occurring at each node. Based on the time series of marginal electricity prices, the electricity price of each node in each time period is determined; Based on the electricity price of each node in each time period, the average electricity price of all nodes in a single time period and the degree of price inequality among all nodes in a single time period are calculated. For each time period, the Gini coefficient for electricity price is determined based on the average electricity price and the degree of electricity price inequality for that time period. The temporal entropy of each node and the Gini coefficient of electricity price for each time period are weighted and fused to determine the spatiotemporal fluctuation index of electricity price for the power grid planning scheme to be evaluated.

[0051] Furthermore, based on the power flow of each line in each time period, the spatial concentration index of congestion costs is determined, including: The marginal electricity price time series of each node is decomposed to extract the congestion price of each line; Based on the predicted active power output of each unit in each time period, the congestion price of each line, and the power flow of each line in each time period, the total line congestion cost of each line in the power grid planning scheme to be evaluated within the time window is obtained. Obtain the power transfer distribution factor between each generating unit and its corresponding lines from the power grid topology parameters; Based on the total line congestion cost of each line within the time window, a congestion cost calculation operation is performed for each unit to obtain the congestion cost of each unit for each line. For each generating unit, the congestion costs of all lines corresponding to the generating unit are summed to obtain the first total congestion cost of each generating unit for all corresponding lines in the power grid planning scheme to be evaluated. Obtain the second total congestion cost for each unit across all corresponding lines in the baseline grid; For each unit, the change in total congestion cost for each unit is obtained based on the second total congestion cost and the first total congestion cost; Based on the change in the total congestion cost of each unit, the spatial concentration index of congestion cost of the power grid planning scheme to be evaluated is obtained. The blocking cost calculation operation includes: Use the current unit as the target unit; The lines corresponding to the target generator set are designated as the target lines. The target generator units and each target line are respectively paired to form several generator-line pairs; For each generator line pair, the congestion cost of the generator line pair is determined based on the total line congestion cost corresponding to the target line in the generator line pair, the power transfer distribution factor corresponding to the generator line pair, and the predicted active power output of the target generator in the generator line pair within the time window, thus obtaining the current unit's congestion cost for the target line.

[0052] Furthermore, based on the predicted active power output of each unit in each time period and the theoretical power generation of new energy sources, the spatiotemporal consumption index of new energy sources is determined, including: The predicted active power output of each unit in each time period is summed to obtain the total new energy output of all units within the time window; The theoretical renewable energy power generation of each unit at each time period is summed to obtain the total theoretical renewable energy output of all units within the time window. The amount of abandoned renewable energy is obtained based on the total output of renewable energy and the theoretical total output of renewable energy. Based on the total output of new energy sources and the amount of new energy curtailment, the initial new energy consumption rate of all units within the time window is obtained; For each time period, the dynamic current carrying capacity of each line in each time period is obtained based on the preset maximum allowable temperature of the line, the standard ambient temperature, the preset meteorological sample data, the preset wind speed correction coefficient, and the static critical value of current carrying capacity under standard conditions. For each line, in each time period corresponding to the current line, the dynamic thermal setpoint of the current line in the current time period is obtained based on the preset three-phase power conversion factor, the dynamic current carrying capacity of the line in the current time period, and the preset rated voltage of the line; the dynamic thermal setpoint is normalized to obtain the normalized thermal setpoint of each line in each time period. The source-load sample data and the preset meteorological sample data are input into the hidden blockage prediction model so that the hidden blockage prediction model can output the instantaneous blockage frequency of each line in each time period. For each time period in each line, the time period in which the instantaneous blocking frequency is greater than the normalized thermal setpoint is selected as the first candidate time period; For each line, the first candidate time periods that are consecutive are connected to obtain several second candidate time periods; the second candidate time periods that are less than the first time threshold are regarded as hidden blocking time periods. The total number of concealed blocking periods across all lines is taken as the concealment count; the duration of each concealed blocking period is taken as the duration of each concealed blocking event. For each hidden blockage period, the reduction in new energy output for each hidden blockage period is determined based on the theoretical power generation of new energy sources corresponding to the hidden blockage period and the predicted active power output of each unit. The correction factor for the initial renewable energy absorption rate is determined based on the number of concealed events, the duration of each concealed blockage event, and the amount of renewable energy output reduction during each concealed blockage period. Based on the correction factor and the initial renewable energy absorption rate, the corrected renewable energy absorption rate is determined. The renewable energy absorption rate is used as the renewable energy spatiotemporal absorption index for the power grid planning scheme to be evaluated.

[0053] Furthermore, based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy sources, the evaluation results of the power grid planning scheme to be evaluated are determined, including: A comprehensive evaluation vector is generated based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal consumption index of new energy sources. Based on the comprehensive evaluation vector and the preset benchmark scheme index vector, the comprehensive deviation degree is obtained; If the overall deviation is less than the preset deviation threshold, the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme is approved. If the overall deviation is greater than or equal to the preset deviation threshold, the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme fails.

[0054] After determining that the evaluation result of the power grid planning scheme to be evaluated is that the scheme is not approved, the following steps are also included: When the electricity price spatiotemporal fluctuation index is greater than the benchmark fluctuation index, the top few nodes with the largest electricity price spatiotemporal fluctuation index are taken as the first target nodes, and new energy storage capacity is added to the first target nodes. When the spatial concentration index of congestion cost is greater than the baseline spatial concentration, the nodes with the largest changes in congestion cost are taken as the second target nodes, and new transmission lines are added to the second target nodes. When the spatiotemporal absorption index of new energy is less than the benchmark spatiotemporal absorption index, the top few lines with the most hidden blockage events are selected as the first target lines, and the transmission capacity of the first target lines is increased.

[0055] In one possible implementation, the temporal entropy of each node is determined as follows: For each node, the marginal electricity price time series corresponding to the node is divided into K price levels (usually K=10), and the probability of each level occurring is calculated. The marginal electricity price time series corresponding to the node can be represented as: (9) Where i represents the node number and T represents the total number of time periods.

[0056] Furthermore, the formula for calculating temporal entropy is: (10) in, For nodes The temporal entropy; For nodes Electricity prices are in the first The probability of each level; Given the number of discretization levels, the range of values ​​for the time series entropy is: The larger the entropy value, the more drastic the fluctuation of electricity prices over time.

[0057] In one possible implementation, the Gini coefficient for electricity prices is determined by sorting the electricity prices of all nodes in ascending order, as shown in the following expression: (11) Where t represents the index of the time period, and N represents the total number of nodes. This indicates the sorted node numbers. Let represent the sorted electricity price. It indicates the electricity price of the k-th node at time t, after sorting in ascending order. This represents the electricity price of the node with the lowest electricity price among all nodes. This represents the electricity price of the node with the highest electricity price among all nodes.

[0058] Furthermore, the Gini coefficient for electricity prices in a single time period is calculated as follows: (12) in, For time period The Gini coefficient for electricity prices, The total number of nodes; and They are nodes and nodes At any moment Electricity price; For a moment The average electricity price. The range of the Gini coefficient for electricity prices is... The larger the value, the more uneven the distribution of electricity prices in the spatial dimension.

[0059] Preferably, the formula for calculating the spatiotemporal fluctuation index of electricity prices for the power grid planning scheme to be evaluated is as follows: (13) in, This refers to the spatiotemporal fluctuation index of electricity prices for the power grid planning scheme to be evaluated. and The weighting coefficients are set, and Default value It can also be adjusted according to the verification objectives; if more attention is paid to the time fluctuation of electricity prices, then the adjustment can be increased. If more attention is paid to spatial distribution, then it increases , Indicates the total number of nodes. Indicates the first The temporal entropy of each node, Indicates the total number of time periods. This represents the Gini coefficient of electricity price for all nodes at the t-th time segment, reflecting the degree of imbalance in the distribution of electricity prices among different nodes at that moment.

[0060] In one possible implementation, after the economic dispatch model is solved, based on the Lagrange duality theory of the model, the Lagrange multipliers corresponding to the power balance constraint, line power flow constraint, and network loss balance constraint are extracted to construct the decomposition calculation formula for the nodal marginal electricity price: (14) in, Let n be the marginal electricity price at node n in time period t. The marginal cost of system electrical energy is the Lagrange multiplier for power balance constraints. For the line The congestion price is a Lagrange multiplier of the line flow constraint. For node n pairs of lines The power transfer distribution factor, The network loss cost is derived from the network loss correction under power balance constraints, and is essentially still derived based on the Lagrange multipliers of power balance constraints.

[0061] In one possible implementation, based on the predicted active power output of each generating unit in each time period, the congestion price of each line, and the power flow of each line in each time period, the total line congestion cost of each line in the power grid planning scheme to be evaluated within the time window is obtained, including: First, based on the predicted active power output of each unit in each time period, the congestion power of each line in each time period t is calculated. Specifically, the formula for calculating the congestion power of each line in each time period t is as follows: (15) in, It is a line The blocking power during time period t, For the unit For the line The contribution coefficient satisfies , For the unit The predicted active power output corresponding to time period t. It should be noted that in formula (15), (unit For the line The contribution coefficient is numerically calculated using the power transfer distribution factor between each generating unit and its corresponding lines. The reasons are as follows: Under the linearized DC power flow model, the power transfer distribution factor essentially represents the linear mapping relationship between the power injected by generating nodes and the power flow in the lines. Its value directly reflects the degree of influence of changes in unit output on the power flow in the lines. The power flow tracing method, based on the proportional allocation principle, performs source decomposition of the power flow in the lines at the same time section. The resulting contribution coefficient corresponds completely to the power transfer distribution factor in both physical meaning and mathematical form, and both satisfy the linear superposition relationship between the power flow in the lines and the generating unit output. Using the power transfer distribution factor as the contribution coefficient allows for the precise quantification of the contribution share of each generating unit's output to the line congestion power, relying on its inherent linear sensitivity characteristics, ensuring the fairness and traceability of subsequent congestion cost allocation. Furthermore, this coefficient matrix is ​​directly derived from the topological relationship of the power flow distribution, eliminating the need for complex nonlinear power flow calculations. It can be directly adapted to the source-fall analysis process based on cross-sectional power flow in this scheme, achieving efficient decomposition from line congestion power to the contribution component of generating nodes.

[0062] The calculated route After the blocking power in time period t, based on the line The congestion cost for each line is obtained by considering the congestion power and congestion price of each line during time period t. The specific calculation formula is as follows: ( in, For each line The cost of blocking For the line The blocking cost at time t For the line The blocking power during time period t.

[0063] In one possible implementation, "obtaining the power transfer distribution factor between each generating unit and its corresponding lines from the grid topology parameters; and performing a congestion cost calculation operation for each generating unit based on the total line congestion cost of each line within the time window, to obtain the congestion cost of each generating unit for each line" specifically includes: the congestion cost calculation operation includes: taking the current generating unit as the target generating unit; taking the several lines corresponding to the target generating unit as target lines; forming several generating unit-line pairs with each target line; and for each generating unit-line pair, executing the following formula: (17) in, This represents the congestion cost of the unit's line pair (i.e., the congestion cost of each unit for each line). Indicates the line The cost of blocking Indicates the unit Corresponding lines The power transfer distribution factor between them For the unit There is always effort and contribution within the time window. The total active power output of each unit within the time window. and They are the same type of variable, only This indicates a traversal of all units. Indicates the unit Corresponding lines The power transfer distribution factor between them.

[0064] In one possible implementation, after obtaining the congestion cost of each unit for each line, the congestion costs of all lines corresponding to each unit are summed to obtain the formula for calculating the first total congestion cost of each unit for all corresponding lines in the power grid planning scheme to be evaluated, as follows: (18) in, This represents the first total congestion cost for each generating unit across all corresponding lines in the power grid planning scheme to be evaluated. This represents the congestion cost of each generator unit for each line.

[0065] In one possible implementation, the specific method for obtaining the second total congestion cost of each generating unit for all corresponding lines in the reference grid is as follows: First, collect the line congestion event records, actual power flow data, and active power output data of each generating unit in the historical operating cycle of the reference grid; second, based on the historical data, statistically analyze the correlation between the output of each generating unit and the congestion events of each line, and quantify the contribution ratio of each generating unit to the congestion of each line; then, according to the total congestion cost of the line in the historical cycle, allocate the cost to each generating unit according to the contribution ratio; finally, sum up all the line congestion costs allocated to the same generating unit to obtain the second total congestion cost of that generating unit for all corresponding lines in the reference grid.

[0066] In one possible implementation, for each unit, the change in total congestion cost for each unit is obtained based on the second total congestion cost and the first total congestion cost. Specifically, (19) in, It is the change in the total blocking cost of each unit. It is the first total congestion cost for each generating unit for all corresponding lines in the power grid planning scheme to be evaluated. It is the second total congestion cost for each unit relative to all corresponding lines in the reference grid.

[0067] In one possible implementation, the spatial concentration index of congestion cost of the power grid planning scheme to be evaluated is obtained based on the change in the total congestion cost of each unit. The specific calculation formula is as follows: (20) in, This represents the change in the total congestion cost of each unit, where N is the total number of units. It is the standard deviation of the change in total congestion cost across all units. It is the mean of the changes in total congestion costs across all units. The smaller the value, the more uniformly the spatial distribution of congestion problems is improved by the power grid planning scheme to be evaluated.

[0068] In one possible implementation, the specific steps for "summing up the theoretical renewable energy power generation of each unit in each time period to obtain the total theoretical renewable energy output of all units within the time window" are as follows: First, based on historical meteorological data of renewable energy power plants, rated capacity of units, and photovoltaic / wind power theoretical output models, calculate the theoretical maximum power generation of each renewable energy unit in each time period without being affected by grid operation constraints, thus obtaining the theoretical renewable energy power generation of each unit in each time period; then, iterate through all time periods within the time window, summing up the theoretical output of all renewable energy units in each time period to obtain the total theoretical renewable energy output of each time period; finally, sum up the total theoretical renewable energy output of all time periods within the time window again to obtain the total theoretical renewable energy output of all units throughout the entire time window.

[0069] In one possible implementation, the specific method for "obtaining the amount of renewable energy wasted based on the total output of renewable energy and the theoretical total output of renewable energy" is as follows: calculate the difference between the theoretical total output of renewable energy and the total output of renewable energy, and this difference is the amount of renewable energy wasted within the time window.

[0070] In one possible implementation, the specific formula for "obtaining the initial renewable energy absorption rate of all units within the time window based on the total renewable energy output and the amount of renewable energy curtailment" is as follows: (twenty one) in, The initial renewable energy consumption rate for all generating units within the time window. For the abandoned electricity of new energy sources, To contribute to the overall development of new energy theory.

[0071] In one possible implementation, the formula for calculating the dynamic current-carrying capacity of each line in each time period is: (twenty two) in, This refers to the dynamic current carrying capacity of each line at each time period. This represents the static critical value of current-carrying capacity under standard conditions. The preset maximum allowable temperature for the circuit. The average ambient temperature during time period t The standard ambient temperature is preferably 40 degrees Celsius. The average wind speed over time period t (unit: m / s). This is the preset wind speed correction factor, where, For the preset coefficient, preferably, It can be set to 0.15. It should be noted that... , It is based on preset meteorological sample data. Specifically, the preset meteorological sample data is obtained as follows: First, historical weather forecast datasets for the past 3-5 years are collected for the areas along the route. The data includes hourly ambient temperature and wind speed forecasts. Second, the historical weather forecast data is preprocessed, including removing missing and outlier values ​​and performing spatial interpolation according to the route's geographical location to obtain the meteorological data sequence corresponding to the route corridor. Subsequently, according to the time granularity required for dynamic current carrying capacity calculation (such as hourly or 15-minute intervals), the preprocessed meteorological data is time-aligned and aggregated to generate the ambient temperature corresponding to each time period. With wind speed Sample; In addition, the method for obtaining the static critical value of current carrying capacity under standard conditions is as follows: First, based on the conductor type, cross-sectional specifications, erection method, and line voltage level of the transmission line, consult the long-term allowable current carrying capacity under the standard environmental conditions (ambient temperature 40℃, no wind, no sunshine) specified in the corresponding national or industry standard (such as "GB / T 1179-2017 Round Wire Concentric Stranded Overhead Conductor") to obtain the theoretical static critical value of current carrying capacity; Second, combine the rated current carrying capacity parameters in the type test report, factory test data, or line design specification before the line is put into operation to verify and correct the theoretical value to ensure that it matches the actual conductor parameters and operating conditions of the line; Finally, the verified current carrying capacity value is taken as the standard static critical value of current carrying capacity for the line.

[0072] In one possible implementation, the covert blocking prediction model is a covert blocking prediction model based on a temporal convolutional neural network (TCN), and the structure of the model is as follows: Input Layer: This layer receives and integrates multi-source time-series feature data, including source load sample data and preset meteorological sample data. The dimension of the input features is C. To capture the temporal correlation of features, the model uses a 7-hour sliding time window to organize the input data into a fixed-length time series, ensuring that the model can simultaneously learn the impact of different features on line congestion. Causal Convolutional Layer: This layer employs a causal convolution mechanism to ensure that when predicting the output at the current moment, the model only uses historical input data from the current moment and earlier, avoiding the leakage of future information and thus strictly guaranteeing the causality of time series prediction. The convolutional kernel size is set to 3, and by extracting local features from the input sequence, it initially captures short-term time series patterns related to line congestion in the feature sequence.

[0073] Dilated Convolutional Layer: This layer employs dilated convolution (attenuated convolution) mechanism, using exponentially increasing dilation coefficients (e.g., 1, 2, 4, 8, …) to effectively expand the receptive field of the convolutional kernel without increasing the number of model parameters or computational complexity. This enables the model to simultaneously capture long-term dependencies in line congestion events, improving its predictive ability for hidden congestion, which has potential temporal correlation characteristics.

[0074] Residual connection layers: Each residual block contains two convolutional layers, sequentially configured with a weight normalization layer, a ReLU activation function, and a Dropout layer (dropout rate set to 0.2). The residual connection structure effectively alleviates the vanishing gradient problem during deep network training, ensuring stable training even after stacking multiple convolutional layers. Weight normalization stabilizes the model training process; the ReLU activation function introduces a non-linear transformation, enhancing the model's expressive power; and the Dropout layer effectively prevents overfitting and improves the model's generalization performance by randomly deactivating some neurons.

[0075] Output Layer: This layer outputs the model's prediction results, specifically the instantaneous congestion frequency of each line within the next hour. The output dimension is L, corresponding one-to-one with the number of lines to be predicted in the power grid. Each line corresponds to a predicted congestion frequency value, directly reflecting the probability level of hidden congestion occurring on that line in the future, providing a basis for subsequent power grid operation control and planning evaluation.

[0076] In one possible implementation, the causal convolutional layer, dilated convolutional layer, residual connection layer and their associated structures (weight normalization, ReLU, Dropout) can be encapsulated in a single residual block. Four residual blocks are set between the input and output layers of the hidden blocking prediction model. The dilation coefficients of the four residual blocks increase exponentially (e.g., 1, 2, 4, 8) to gradually expand the model's receptive field and better capture long-term dependencies.

[0077] In one possible implementation, the training sample set for the hidden congestion prediction model is obtained by extracting source-load sample data and historical meteorological data from the power grid historical database for the past three consecutive years, with a data sampling frequency of once per hour. Each 72-hour time window is used as a sample, and the corresponding label is whether each line will experience congestion in the next hour (binary classification label). In one example, approximately 26,000 samples are obtained. The samples are divided into training set, validation set, and test set in a 7:2:1 ratio. The data dimension input to the hidden congestion prediction model is 72×C, where C is the feature dimension, including: power output, load, temperature, wind speed, and humidity, a total of 5 dimensions.

[0078] In one possible implementation, the training process for the covert blockage prediction model is as follows: using source load sample data from the past three years and pre-set meteorological sample data as the training set, and employing binary cross-entropy loss as the loss function: (twenty three) in, The total number of samples, For real labels, For predicting labels, It is a sample index.

[0079] Preferably, the optimizer of the model is Adam with a learning rate of 0.001. The validation set consists of source load sample data from the most recent 3 months and preset meteorological sample data. Training stops when the validation set loss does not decrease for 10 consecutive rounds.

[0080] In one possible implementation, the Adam optimizer is used with an initial learning rate of 0.001, which decays to 0.9 times the original rate every 10 epochs. The loss function is a binary cross-entropy loss, with L2 regularization (regularization coefficient 0.0001) added to prevent overfitting. Training stops when the validation set loss does not decrease for 10 consecutive epochs. Through this training, the TCN model can accurately capture the temporal dependency between new energy fluctuations and instantaneous congestion frequency, providing a basis for early warning of hidden congestion events.

[0081] In one possible implementation, for each line, in each time period corresponding to the current line, the dynamic thermal setpoint of the current line in the current time period is obtained based on the preset three-phase power conversion factor, the dynamic current carrying capacity of the line in the current time period, and the preset rated voltage of the line. This includes: for each line, in each time period corresponding to the current line, multiplying the preset three-phase power conversion factor, the dynamic current carrying capacity of the line in the current time period, and the preset rated voltage of the line to obtain the dynamic thermal setpoint of the current line in the current time period.

[0082] It should be noted that the method for determining the preset rated voltage of the line is as follows: the standard voltage level marked in the engineering design drawings and project planning documents of a single transmission line is preset, and each line is matched with its own corresponding voltage level, such as 10kV, 110kV, 220kV, 500kV, etc. Different lines can be configured with different rated voltages.

[0083] In one possible implementation, the dynamic thermal setpoint is normalized to obtain the normalized thermal setpoint for each line at each time period, including: normalizing the preset static critical current carrying capacity of the line. Multiply the dynamic thermal setpoint by the preset rated voltage of the line to obtain the reference maximum power of the line; then divide the dynamic thermal setpoint by the reference maximum power of the line to obtain the normalized thermal setpoint.

[0084] It should be noted that the preset three-phase power conversion factor can be taken as... .

[0085] It should be noted that the preset static critical current for each line is set as follows: based on the conductor type, cross-section and voltage level of each line, the long-term allowable current carrying capacity of the conductor under standard conditions is obtained by referring to the power industry standards, and this is directly used as the preset static critical current for that line. Different lines are configured independently, and the preset static critical current values ​​for different lines may be different.

[0086] In one possible implementation, the specific implementation of "for each time period in each line, the time period with an instantaneous blocking frequency greater than the normalized thermal setpoint is taken as the first candidate time period; for each line, the consecutive first candidate time periods are connected to obtain several second candidate time periods; the second candidate time periods with a frequency less than the first time threshold are taken as the hidden blocking time period" is as follows: when the following conditions are met, the corresponding time interval of the line is identified as the hidden blocking time period: for a certain line, there exists a time interval , making For all Established; among them, For instantaneous blocking frequency, For normalized heat constant, time interval satisfy , The first time threshold is the minimum start-up and shutdown time of the unit. Preferably, the minimum start-up and shutdown time of the unit is 60 to 120 minutes.

[0087] In one possible implementation, the time interval must also meet the following condition to be identified as a hidden blocking period: within the time interval During this period, the threshold for renewable energy power carried by the affected transmission sections associated with the line must be greater than the preset scale threshold, i.e., it must meet the following formula: (twenty four) in, The threshold for renewable energy power carried by the affected transmission sections associated with the line. These represent the minimum and maximum installed capacities of the renewable energy units connected to the transmission section, respectively. 10MW is a preset benchmark coefficient based on grid operation experience, used to convert the installed capacity into an engineering-significant power threshold value. It should be noted that this refers to the renewable energy power threshold carried by the affected transmission section. It represents the sum of the actual grid-connected power output of all new energy units (including wind power and photovoltaic) connected to the transmission section within the time interval, and is used to quantify the direct impact of the congestion event on the consumption of new energy.

[0088] In one possible implementation, the specific steps for "determining the reduction in new energy output for each hidden congestion period based on the theoretical power generation of new energy sources and the predicted active power output of each generating unit corresponding to the hidden congestion period" are as follows: First, for each identified hidden congestion period, extract the theoretical output sequence of all new energy power plants connected to the affected transmission section within that period, i.e., the maximum power generation capacity without considering grid operation constraints; second, obtain the predicted active power output sequence of each new energy generating unit within the same period, i.e., the actual grid-connected power after considering line congestion constraints; subsequently, calculate the difference between the theoretical power generation of new energy sources and the predicted active power output within that period, which is the reduction in new energy output caused by hidden congestion.

[0089] In one possible implementation, the specific calculation formula for "determining the correction factor for the initial renewable energy absorption rate based on the number of concealment events, the duration of each concealment blockage event, and the renewable energy output reduction during each concealment blockage period" is as follows: (25) in, This is a correction factor for the initial renewable energy absorption rate. For the purpose of concealing the number of times, For the first The duration of each hidden blocking period For the first The reduction in new energy output during periods of hidden congestion. For the first The line number where the incident occurred. For duration weighting function, For the reduction amount weight function, This is the importance weighting function for the routes. Specifically, It is an S-shaped function, specifically: (26) in, For the first The duration of each hidden blocking period This is the median threshold for the duration of the covert blocking period, with a default value of 15 minutes. This is the steepness parameter for the S-curve, with a default value of 0.5.

[0090] It is a piecewise linear function, specifically: (27) in, For the first The reduction in new energy output during periods of hidden congestion. This is the minimum threshold for peak shaving capacity from new energy sources, with a default value of 10MWh. This is the saturation threshold, with a default value of 100MWh.

[0091] The specific formula is: (28) in, For nodes To the node The total number of shortest paths between nodes (for example, if there are 3 shortest paths of the same length from node A to node B, then...). ), For nodes To the node Of all the shortest paths, the one that passes through the line The number of paths, For the first The line number where the incident occurred. Characterization circuit For nodes To the node The passage weight, that is, from arrive We must go The probability of.

[0092] Preferably, the weighting function in the correction factor λ , The parameters are not fixed, but are adaptively optimized based on historical data and operational experience. Specifically, this includes: 1. Collecting hidden congestion periods from historical records and their corresponding actual absorption indices to construct an optimization sample set; 2. Using Bayesian optimization methods to minimize the error between the corrected spatiotemporal absorption index and the actual absorption index. In , ,and In , Parameter optimization is performed, and the objective function is optimized as follows: (29) Where M is the total number of samples in the hidden blocking period of the optimized sample set. This is the corrected spatiotemporal absorption index of new energy calculated using the current parameters. 3. Output the optimized parameters: (This refers to the corresponding actual absorption index.) , , , This is used for subsequent correction factor calculations. The correction factor obtained in this way can more accurately reflect the actual impact of the hidden blockage period on the absorption of new energy sources, further improving the reliability of the verification results.

[0093] In one possible implementation, the formula for "determining the corrected renewable energy absorption rate based on the correction factor and the initial renewable energy absorption rate" is as follows: (30) in, This is the revised renewable energy consumption rate. As a correction factor, This represents the initial renewable energy consumption rate.

[0094] In one possible implementation, "generating a comprehensive evaluation vector based on the spatiotemporal fluctuation index of electricity price, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy" includes: generating the following vector from the spatiotemporal fluctuation index of electricity price, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy: ,in, This represents the spatiotemporal fluctuation index of electricity prices. This indicates the spatial concentration of congestion costs. This represents the spatiotemporal consumption index of new energy.

[0095] In one possible implementation, "obtaining the comprehensive deviation based on the comprehensive evaluation vector and the preset benchmark scheme index vector" includes: the calculation formula for the comprehensive deviation is as follows: (31) in, For the comprehensive deviation, This is the index of spatiotemporal fluctuations in electricity prices. This is a preset benchmark indicator vector. To block the spatial concentration index of cost. The index for the spatiotemporal absorption of new energy. The weighting coefficients are preset; preferably, the weighting for new energy consumption is slightly higher. .

[0096] Preferably, "if the overall deviation is less than the preset deviation threshold, then the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme is approved" includes: the preset deviation threshold is 0.1.

[0097] In one possible implementation, when the spatiotemporal volatility index of electricity prices is greater than the benchmark volatility index (i.e., When the value of the power grid planning scheme to be evaluated is not good, it indicates that the average effect of the power price spatiotemporal fluctuation is not good. At this time, the following suggestions are given for the configuration of the new energy storage system: take the top few nodes with the largest power price spatiotemporal fluctuation index as the first target nodes, and add new energy storage capacity to the first target nodes. Preferably, the new energy storage capacity adopts batteries with fast response speed.

[0098] In one possible implementation, when the spatial concentration index of blocking costs is greater than the baseline spatial concentration ( When the situation indicates that the planning scheme does not improve the spatial distribution of congestion problems evenly, and there are units with significantly increased congestion costs, the output is: select the top few units with the largest changes in congestion costs as target units, and add new transmission lines to the nodes where the target units are located; specifically, firstly, select the units with positive changes in congestion costs and ranked in the top N, and the access nodes corresponding to these units are the priority access nodes for new lines. These are the locations where the increase in congestion costs is most concentrated under the current planning scheme. Then, based on the total congestion cost allocated to the target unit nodes and the renewable energy output scale of the node, determine the rated capacity of the new transmission lines, so that the transmission capacity of the lines is commensurate with the congestion of the node. The cost level is matched to ensure that the new lines can effectively channel the surplus output of the target units and reduce congestion pressure. Based on the power grid topology and the power flow direction of the target unit nodes, the optimal path of the new transmission lines is determined by the topology optimization algorithm. Priority is given to the route of the line that can open up the power transmission bottleneck while taking into account other potential congestion nodes, so as to maximize the reduction of the spatial concentration of the overall network congestion cost. Through the above steps, the new transmission lines can directly reinforce the unit nodes where the congestion cost increases most significantly, effectively alleviate the new uneven distribution of congestion after the implementation of the planning scheme, thereby reducing the spatial concentration index of congestion cost and improving the overall balance and security of the power grid operation.

[0099] When the spatiotemporal absorption index of new energy is less than the benchmark spatiotemporal absorption index (i.e.) When the situation is as follows, it indicates that the planning scheme is insufficient in dealing with hidden congestion. At this time, the mitigation suggestions for hidden congestion events are output: take the top few lines with the most hidden congestion events as the first target lines, and increase the transmission capacity of the first target lines. "Increasing the transmission capacity of the lines" refers to increasing the upper limit of the dynamic / static current carrying capacity of the lines through technical transformation or capacity expansion measures. Specifically, it can be done by replacing the conductors with larger cross-section conductors, adding series compensation devices, using new heat-resistant conductors or dynamic capacity expansion systems, etc., to directly improve the safe transmission capacity of the lines, reduce hidden congestion events caused by line overload, and thus improve the bottleneck of new energy consumption.

[0100] like Figure 2 As shown, based on the above method embodiments, corresponding apparatus embodiments are provided; An embodiment of the present invention provides an evaluation device for power grid planning schemes, comprising: a historical source-load data determination module, a parameter determination module, a source-load sample data determination module, a unit combination model determination module, a first solution module, an economic dispatch model determination module, a second solution module, an evaluation parameter determination module, and an evaluation module; The historical source load data determination module is used to obtain historical source load data of the reference power grid; The parameter determination module is used to obtain the total number of generating units, the length of the time window, the operating constraint parameters of each generating unit, the cost parameters of each generating unit, the grid topology parameters, and the theoretical power generation of each generating unit in each time period for the power grid planning scheme to be evaluated. The source load sample data determination module is used to obtain source load sample data for each time period in the time window based on historical source load data. The unit combination model determination module is used to construct a unit combination model based on the cost parameters of each unit and the length of the time window, with the goal of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and to construct the first constraint condition based on the operating constraint parameters of each unit, the power grid topology parameters, the source-load sample data, and the length of the time window. The first solution module is used to solve the unit combination model under the first constraint condition to obtain the start-up and shutdown status of each unit in each time period; The economic dispatch model determination module is used to construct an economic dispatch model based on the total number of generating units, the length of the time window, and the preset fitting coefficients for the output cost of each generating unit, with the goal of minimizing the total operating cost of the power grid planning scheme to be evaluated within the time window. Based on the start-up and shutdown status, the module updates the first constraint and generates the second constraint. The second solution module is used to solve the economic dispatch model under the second constraint to obtain the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. The evaluation module is used to determine the evaluation results of the power grid planning scheme to be evaluated based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period.

[0101] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the evaluation method for power grid planning schemes provided by any of the above-described method embodiments of the present invention.

[0102] It should be noted that the device embodiments described above are merely illustrative, and some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Furthermore, in the accompanying drawings of the device embodiments provided by this invention, the connection relationships between modules indicate that they have communication connections, which can specifically be implemented as one or more communication buses or signal lines. Those skilled in the art can understand and implement this without any creative effort.

[0103] Based on the above-described embodiments of the power grid planning scheme evaluation method, another embodiment of the present invention provides a terminal device, which includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, it implements the power grid planning scheme evaluation method of any embodiment of the present invention.

[0104] For example, in this embodiment, the computer program can be divided into one or more modules, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules may be a series of computer program instruction segments capable of performing a specific function, which describe the execution process of the computer program in the terminal device.

[0105] The terminal device may be a desktop computer, laptop, handheld computer, or cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory.

[0106] The processor can be a Central Processing Unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor can be a microprocessor or any conventional processor. The processor is the control center of the terminal device, connecting all parts of the terminal device via various interfaces and lines.

[0107] Based on the above-described method embodiments, another embodiment of the present invention provides a computer-readable storage medium including a stored computer program, wherein, when the computer program is executed, it controls the device where the computer-readable storage medium is located to execute the power grid planning scheme evaluation method described in any of the above-described method embodiments of the present invention.

[0108] The modules / units integrated in the device / terminal equipment, if implemented as software functional units and sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, all or part of the processes in the above embodiments of the present invention can also be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium can include: any entity or device capable of carrying the computer program code, a recording medium, a USB flash drive, a portable hard drive, a magnetic disk, an optical disk, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium, etc.

[0109] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.

Claims

1. A method for evaluating power grid planning schemes, characterized in that, include: Obtain historical source-load data from the benchmark power grid; Obtain the total number of generating units, the length of the time window, the operating constraints of each generating unit, the cost parameters of each generating unit, the grid topology parameters, and the theoretical power generation of each generating unit in each time period for the power grid planning scheme to be evaluated; Based on historical source load data, source load sample data for each time period within the time window are obtained; Based on the cost parameters of each generating unit and the length of the time window, a generating unit combination model is constructed with the goal of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and based on the operating constraint parameters of each generating unit, the power grid topology parameters, the source-load sample data, and the length of the time window, the first constraint condition is constructed. Under the first constraint, the unit combination model is solved to obtain the start-up and shutdown status of each unit in each time period, the maximum output limit and the minimum output limit of each unit in each time period. Based on the total number of generating units, the length of the time window, and the preset fitting coefficients for the output cost of each generating unit, an economic dispatch model is constructed with the goal of minimizing the total operating cost of the power grid planning scheme to be evaluated within the time window. Based on the start-up and shutdown status, the maximum and minimum output limits of each generating unit in each time period, the first constraint is updated to generate the second constraint. Under the second constraint, the economic dispatch model is solved to obtain the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. Based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period, the evaluation results of the power grid planning scheme to be evaluated are determined.

2. The evaluation method for power grid planning schemes as described in claim 1, characterized in that, Historical source-load data includes: historical renewable energy output data and historical load data; Based on historical source-load data, source-load sample data for each time period within the time window are obtained, including: Feature extraction was performed on historical renewable energy output data to obtain historical output characteristics including daily maximum output, daily minimum output, daily average output volatility, and daily peak-to-valley difference. Feature extraction is performed on historical load data to obtain historical load characteristics including daily maximum load, daily minimum load, daily average load fluctuation rate, and daily load peak-to-valley difference; Based on historical power output characteristics and historical load characteristics, historical renewable energy power output data and historical load data are classified into several typical scenarios. For each typical scenario, several source-load sample data corresponding to the typical scenario are generated to obtain the source-load sample data for each time period in the time window.

3. The evaluation method for power grid planning schemes as described in claim 2, characterized in that, The cost parameters for each unit include: start-up and shutdown costs, operating cost coefficients, and fixed operating costs for each unit. The unit combination model includes: in, The total number of units. For the unit's index, For time period index, The length of the time window, For the unit Start-up and shutdown costs For the unit During the period Start-stop status, For the unit During the period Start-stop status, Let g be the operating cost coefficient of unit g. For the unit During the period The predictions were helpful and productive. unit Fixed operating costs.

4. The evaluation method for power grid planning schemes as described in claim 3, characterized in that, The economic scheduling model includes: in, The total number of units. For the unit's index, For time period index, The length of the time window, , , The fitting coefficients for the output cost of each unit. For the unit During the period The predictions were helpful and productive.

5. The evaluation method for power grid planning schemes as described in claim 4, characterized in that, Based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period, the evaluation results of the power grid planning scheme to be evaluated are determined, including: Based on the marginal electricity price time series, the spatiotemporal fluctuation index of electricity price is determined; based on the power flow of each line in each time period, the spatial concentration index of congestion cost is determined; based on the predicted active power output of each unit in each time period and the theoretical power generation of new energy, the spatiotemporal consumption index of new energy is determined. The evaluation results of the power grid planning scheme to be evaluated are determined based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy sources.

6. The evaluation method for power grid planning schemes as described in claim 5, characterized in that, The determination of the spatiotemporal fluctuation index of electricity prices based on the marginal electricity price time series includes: For each node, the marginal electricity price time series corresponding to the node is divided into several price levels, and the probability of each price level occurring is calculated. The temporal entropy of each node is obtained based on the probability of each electricity price level occurring at each node. Based on the time series of marginal electricity prices, the electricity price of each node in each time period is determined; Based on the electricity price of each node in each time period, the average electricity price of all nodes in a single time period and the degree of price inequality among all nodes in a single time period are calculated. For each time period, the Gini coefficient for electricity price is determined based on the average electricity price and the degree of electricity price inequality for that time period. The temporal entropy of each node and the Gini coefficient of electricity price for each time period are weighted and fused to determine the spatiotemporal fluctuation index of electricity price for the power grid planning scheme to be evaluated.

7. The evaluation method for power grid planning schemes as described in claim 6, characterized in that, Based on the power flow of each line in each time period, the spatial concentration index of congestion costs is determined, including: The marginal electricity price time series of each node is decomposed to extract the congestion price of each line; Based on the predicted active power output of each unit in each time period, the congestion price of each line, and the power flow of each line in each time period, the total line congestion cost of each line in the power grid planning scheme to be evaluated within the time window is obtained. Obtain the power transfer distribution factor between each generating unit and its corresponding lines from the power grid topology parameters; Based on the total line congestion cost of each line within the time window, a congestion cost calculation operation is performed for each unit to obtain the congestion cost of each unit for each line. For each generating unit, the congestion costs of all lines corresponding to the generating unit are summed to obtain the first total congestion cost of each generating unit for all corresponding lines in the power grid planning scheme to be evaluated. Obtain the second total congestion cost for each unit across all corresponding lines in the baseline grid; For each unit, the change in total congestion cost for each unit is obtained based on the second total congestion cost and the first total congestion cost; Based on the change in the total congestion cost of each unit, the spatial concentration index of congestion cost of the power grid planning scheme to be evaluated is obtained. The blocking cost calculation operation includes: Use the current unit as the target unit; The lines corresponding to the target generator set are designated as the target lines. The target generator units and each target line are respectively paired to form several generator-line pairs; For each generator line pair, the congestion cost of the generator line pair is determined based on the total line congestion cost corresponding to the target line in the generator line pair, the power transfer distribution factor corresponding to the generator line pair, and the predicted active power output of the target generator in the generator line pair within the time window, thus obtaining the current unit's congestion cost for the target line.

8. The evaluation method for power grid planning schemes as described in claim 7, characterized in that, The spatiotemporal absorption index of new energy is determined based on the predicted active power output of each unit in each time period and the theoretical power generation of new energy sources, including: The predicted active power output of each unit in each time period is summed to obtain the total new energy output of all units within the time window; The theoretical renewable energy power generation of each unit at each time period is summed to obtain the total theoretical renewable energy output of all units within the time window. The amount of abandoned renewable energy is obtained based on the total output of renewable energy and the theoretical total output of renewable energy. Based on the total output of new energy sources and the amount of new energy curtailment, the initial new energy consumption rate of all units within the time window is obtained; For each time period, the dynamic current carrying capacity of each line in each time period is obtained based on the preset maximum allowable temperature of the line, the standard ambient temperature, the preset meteorological sample data, the preset wind speed correction coefficient, and the static critical value of current carrying capacity under standard conditions. For each line, in each time period corresponding to the current line, the dynamic thermal setpoint of the current line in the current time period is obtained based on the preset three-phase power conversion factor, the dynamic current carrying capacity of the line in the current time period, and the preset rated voltage of the line; the dynamic thermal setpoint is normalized to obtain the normalized thermal setpoint of each line in each time period. The source-load sample data and the preset meteorological sample data are input into the hidden blockage prediction model so that the hidden blockage prediction model can output the instantaneous blockage frequency of each line in each time period. For each time period in each line, the time period in which the instantaneous blocking frequency is greater than the normalized thermal setpoint is selected as the first candidate time period; For each line, the first candidate time periods that are consecutive are connected to obtain several second candidate time periods; the second candidate time periods that are less than the first time threshold are regarded as hidden blocking time periods. The total number of concealed blocking periods across all lines is taken as the concealment count; the duration of each concealed blocking period is taken as the duration of each concealed blocking event. For each hidden blockage period, the reduction in new energy output for each hidden blockage period is determined based on the theoretical power generation of new energy sources corresponding to the hidden blockage period and the predicted active power output of each unit. The correction factor for the initial renewable energy absorption rate is determined based on the number of concealed events, the duration of each concealed blockage event, and the amount of renewable energy output reduction during each concealed blockage period. Based on the correction factor and the initial renewable energy absorption rate, the corrected renewable energy absorption rate is determined. The renewable energy absorption rate is used as the renewable energy spatiotemporal absorption index for the power grid planning scheme to be evaluated.

9. The evaluation method for power grid planning schemes as described in claim 8, characterized in that, Based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal absorption index of new energy sources, the evaluation results of the power grid planning scheme to be evaluated are determined, including: A comprehensive evaluation vector is generated based on the spatiotemporal fluctuation index of electricity prices, the spatial concentration index of congestion costs, and the spatiotemporal consumption index of new energy sources. Based on the comprehensive evaluation vector and the preset benchmark scheme index vector, the comprehensive deviation degree is obtained; If the overall deviation is less than the preset deviation threshold, the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme is approved. If the overall deviation is greater than or equal to the preset deviation threshold, the evaluation result of the power grid planning scheme to be evaluated is determined to be that the scheme fails.

10. An evaluation device for power grid planning schemes, characterized in that, include: The system includes a historical source load data determination module, a parameter determination module, a source load sample data determination module, a unit combination model determination module, a first solution module, an economic dispatch model determination module, a second solution module, and an evaluation module. The historical source load data determination module is used to obtain historical source load data of the reference power grid; The parameter determination module is used to obtain the total number of generating units, the length of the time window, the operating constraint parameters of each generating unit, the cost parameters of each generating unit, the grid topology parameters, and the theoretical power generation of each generating unit in each time period for the power grid planning scheme to be evaluated. The source load sample data determination module is used to obtain source load sample data for each time period in the time window based on historical source load data. The unit combination model determination module is used to construct a unit combination model based on the cost parameters of each unit and the length of the time window, with the goal of minimizing the total start-up and shutdown costs and total operating costs of the power grid planning scheme to be evaluated within the time window; and to construct the first constraint condition based on the operating constraint parameters of each unit, the power grid topology parameters, the source-load sample data, and the length of the time window. The first solution module is used to solve the unit combination model under the first constraint conditions to obtain the start-up and shutdown status of each unit in each time period, the maximum output limit and the minimum output limit of each unit in each time period. The economic dispatch model determination module is used to construct an economic dispatch model based on the total number of generating units, the length of the time window, and the preset fitting coefficients of the output cost of each generating unit, with the goal of minimizing the total operating cost of the power grid planning scheme to be evaluated within the time window. It also updates the first constraint based on the start-up and shutdown status, the maximum output limit and the minimum output limit of each generating unit in each time period, and generates the second constraint condition. The second solution module is used to solve the economic dispatch model under the second constraint to obtain the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period. The evaluation module is used to determine the evaluation results of the power grid planning scheme to be evaluated based on the marginal electricity price time series of each node, the predicted active power output of each unit in each time period, and the power flow of each line in each time period.

Images