AC / DC hybrid power grid partitioning method and system considering modularity and reactive power margin

Abstract

The invention discloses an AC-DC hybrid power grid partitioning method and system considering modularity and reactive margin, and the method comprises the steps: extracting nodes in an AC-DC hybrid power grid, and constructing an adjacent matrix of the AC-DC hybrid power grid; generating a plurality of initial partition schemes based on the communication condition of the nodes in the adjacent matrix; calculating a comprehensive weight index for the initial partitioning scheme to serve as a fitness value; based on the fitness value, screening to obtain a plurality of optimized partition schemes; according to the fitness value of the optimized partition scheme, corresponding crossover probability and mutation probability are given; and continuously iterating, and obtaining a final partitioning scheme after the maximum number of iterations is reached. The obtained partitioning result can effectively improve the voltage level of the alternating-current and direct-current hybrid power grid, the cross-regional active and reactive power transmission capacity is reduced, and theoretical support is provided for reactive power optimization and voltage control of the power grid.

Description

Technical Field

[0001] This invention belongs to the field of AC / DC hybrid power grid zoning technology, and particularly relates to a method and system for AC / DC hybrid power grid zoning that considers modularity and reactive power margin. Background Technology

[0002] Multi-infeed DC transmission systems can expand system transmission capacity and increase operational flexibility, but they can also cause adverse interactions between AC / DC systems and between DC systems. The increase in DC transmission capacity weakens the receiving-end AC system, highlighting the "strong DC, weak AC" contradiction. If the receiving-end AC system experiences voltage anomalies or faults, it can lead to commutation failures at inverter stations, and in severe cases, even interruptions in multiple DC power transmissions, ultimately threatening the entire power system. Therefore, accurately and quickly identifying weak areas in the AC grid is crucial for ensuring the safe and stable operation of large-scale AC / DC grids. Existing technologies, such as the patent document with publication number CN108233359A, provide a partitioning optimization method and system for ultra-high voltage AC / DC infeed receiving-end grids. This method simplifies the power network by abstracting all buses in the power network as undifferentiated nodes and all transmission lines and transformer branches as weighted edges in an undirected network, with the weight being the admittance modulus of the branch. The power network is abstracted as an undirected, weighted sparsely connected graph with n nodes, represented by an n*n order correlation matrix A. After simplifying the power network, the set of critical power grid channels is determined. The optimal partitioning scheme is obtained by iterative multi-objective optimization of randomly generated partitioning schemes using the NSGA-II non-dominated genetic algorithm based on fast classification. Existing AC / DC hybrid power grid partitioning methods mostly focus on network topology optimization, failing to fully consider the response time and regulation capacity differences of reactive power regulation equipment within the region, leading to a disconnect between the partitioning results and actual voltage control requirements.

[0003] Therefore, there is an urgent need for a partitioning method that takes into account both tight coupling within partitions and sparse coupling between regions, while also taking into account the autonomy of reactive power regions, in order to ensure the safe and stable operation of large-scale AC / DC hybrid power grids. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention proposes a partitioning method for AC / DC hybrid power grids that considers regional modularity and reactive power regulation margin. This method introduces modularity and reactive power regulation margin to form a comprehensive weight index, considering regional structural strength and the rational distribution of reactive power regulation resources to obtain an optimized partitioning scheme for the AC / DC hybrid power grid. The modularity index in the proposed method describes the topology and node coupling relationship of the AC / DC hybrid power grid, while the regional voltage regulation capability index mainly describes the response time and regulation capacity of reactive power regulation equipment. The method adheres to the principle of tight coupling within partitions and sparse coupling between regions, while also considering regional autonomy in reactive power regulation. The comprehensive weight of the index is determined using order relation analysis and inverse entropy weighting. Finally, an improved genetic algorithm is used to solve the problem. The resulting partitioning can effectively improve the voltage level of the AC / DC hybrid power grid, reduce the cross-regional active and reactive power transmission capacity, and provide theoretical support for power grid reactive power optimization and voltage control.

[0005] Compared to traditional methods, this approach prioritizes tight coupling within each zone and sparse coupling between zones. It introduces a modularity index to describe the coupling relationship between the AC / DC hybrid power grid topology and nodes. Unlike traditional methods that require pre-setting the number of zones, this modularity index automatically generates the optimal number of zones based on the grid topology and node coupling relationship, avoiding arbitrary zone designation. The response time and regulation capacity of reactive power regulation equipment are incorporated into the zone voltage regulation capability index. Static reactive power regulation resources, such as fixed switching capacitor banks, are prioritized to provide basic reactive power, while the remainder is supplemented by dynamic reactive power regulation resources such as SVG (Static Var Generator). This ensures the system has a certain reactive power regulation margin to cope with system faults or load surges, achieving intra-zone autonomy and guaranteeing the safe and stable operation of the system.

[0006] The present invention adopts the following technical solution.

[0007] Zoning methods for AC / DC hybrid power grids considering modularity and reactive power margin include: Extract the nodes in the AC / DC hybrid power grid and construct the adjacency matrix of the AC / DC hybrid power grid; based on the connectivity of the nodes in the adjacency matrix, generate several initial partitioning schemes; A comprehensive weight index is calculated for the initial partitioning scheme, which is used as the fitness value; based on the fitness value, several optimized partitioning schemes are selected. Based on the fitness value of the optimized partitioning scheme, assign corresponding crossover and mutation probabilities; iterate continuously until the maximum number of iterations is reached to obtain the final partitioning scheme; The comprehensive weighting index is related to the modularity index and the regional voltage regulation capability index of the optimized zoning scheme; The modularity index is related to the electrical distance between different nodes in the adjacency matrix; the regional voltage regulation capability index is the average of the regional voltage regulation capability index theoretically achieved by all partitions in the optimized partitioning scheme; To ensure that each zone has the theoretical capability to regulate the regional voltage, the output sequence and regulation range of the reactive power regulation equipment within the zone are set.

[0008] More preferably, the modularity index and the average value of voltage regulation capability index for all regions Each is assigned a weight coefficient, and the comprehensive weight index γ is a modularity index. and the average value of voltage regulation capability index for all regions The weighted sum of the weights, where the sum of the weight coefficients is 1.

[0009] More preferably, the modularity index The calculation method is as follows: Calculate the voltage change at node i corresponding to the unit value of reactive power change at node j, and obtain the voltage sensitivity matrix to reactive power. Using the element in the j-th row and j-th column of the sensitivity matrix as the numerator and the element in the ith row and j-th column as the denominator, the ratio is obtained. Taking the logarithm of the ratio gives the influence degree of node j on node i. Based on the influence of node j on node i, the electrical distance between node i and node j is calculated; the electrical distance between node i and node j is then converted into inter-node edge weights. The modularity index is calculated using the edge weights between the nodes. .

[0010] More preferably, the electrical distance and the edge weights between nodes are calculated as follows: In a hybrid AC / DC power grid, the influence of any node on node i is subtracted from the influence of any node on node j, resulting in n differences. The electrical distance between node i and node j is obtained by squaring all the differences, summing them, and then taking the square root. To satisfy the relationship between edge weights and electrical distances between nodes, i.e., the smaller the electrical distance, the larger the edge weight, for the electrical distance between nodes i and j, the normalized electrical distance is subtracted from 1 to obtain the edge weights between nodes; Calculate the difference between the edge weights between nodes and the expected edge weights; sum all the differences and divide by twice the total edge weights to obtain the modularity index. .

[0011] More preferably, for a certain zone, its regional voltage regulation capability index is related to the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the zone and the voltage regulation capability of the active power of the photovoltaic unit. The voltage regulation capability of photovoltaic power generation or reactive power compensation equipment in the zone is summed with the voltage regulation capability of active power of photovoltaic units to obtain a summation result; the summation result is compared with 1, and the smaller one is taken as the regional voltage regulation capability index of the zone.

[0012] More preferably, the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition is calculated as follows: compute nodes The reactive power compensation capacity of adjustable resources and nodes Unit value of reactive power change corresponding to node The first product of the voltage changes is summed with the first products corresponding to all nodes in the partition to obtain a first threshold; when the node with the largest voltage deviation in the partition... If the voltage deviation is less than or equal to the first threshold, then the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition is equal to 1; otherwise, the first threshold is used as the numerator, and the node with the largest voltage deviation in the partition is selected. The voltage limit is used as the denominator to obtain the first ratio. The first ratio of all nodes in the partition is summed to obtain the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition.

[0013] More preferably, the method for calculating the voltage regulation capability of the active power of the photovoltaic units in the partition is as follows: compute nodes The maximum active power capacity of adjustable resources and nodes Unit value of active power change corresponding to node The second product of the voltage change values ​​is summed with the second products corresponding to all nodes in the partition to obtain the second threshold; when the node with the largest voltage deviation in the partition... If the voltage deviation is less than or equal to the second threshold, then the voltage regulation capability of the photovoltaic units in the partition is equal to 1; otherwise, the second threshold is used as the numerator, and the node with the largest voltage deviation in the partition is selected. The voltage limit is used as the denominator to obtain the second ratio. The second ratios of all nodes in the partition are summed to obtain the voltage regulation capability of the active power of the photovoltaic unit in the partition.

[0014] More preferably, the step of setting the output sequence of the reactive power regulating devices and the adjustment range of the reactive power regulating devices within the partition specifically involves: Optimize the output of reactive power regulation equipment based on the adjustment response time, prioritizing the adjustment of dynamic reactive power photovoltaic power generation and SVG, and then adjusting static reactive power capacitor banks; The intersection of capacity and power factor constraints is used as the reactive power regulation range for photovoltaic power generation. The square root of the difference between the photovoltaic power generation capacity and the active power of photovoltaic power generation at node i is taken as the upper limit threshold of the reactive power of photovoltaic power generation at node i; a negative sign is added to the square root result as the lower limit threshold of the reactive power of photovoltaic power generation at node i. The power factor of distributed photovoltaic power is between the lower limit of the power factor and 1; The sum of the reactive power adjustment margins and the sum of the reactive power adjustment margins of SVG satisfy the following: The difference between the adjustable upper limit and the real-time output of the i-th SVG reactive power is calculated, and the difference results of all SVGs are summed. The sum is used as the total upper limit adjustment margin of SVG reactive power. The difference between the real-time output and the adjustable lower limit of the i-th SVG reactive power is calculated, and the difference results of all SVGs are summed. The sum is used as the total lower limit adjustment margin of SVG reactive power. The connected capacity of the capacitor bank at node i is between the upper and lower limits of the connected capacity. The optimal operating point of the i-th SVG satisfies the following condition: at the optimal operating point, the output value of the i-th SVG is the average of the adjustable upper limit and the adjustable lower limit of the reactive power of the i-th SVG.

[0015] More preferably, a comprehensive weight index is calculated for the initial partitioning scheme as a fitness value; based on the fitness value, several optimized partitioning schemes are selected and subjected to crossover and mutation. Based on the fitness value of the optimized partitioning scheme, assign corresponding crossover and mutation probabilities; iterate continuously until convergence to obtain the final partitioning scheme; Set corresponding maximum and minimum values ​​for the crossover probability and the mutation probability, respectively; In each iteration, if the fitness value of the two individuals undergoing crossover is greater than the average fitness value of the population, the difference between the maximum and minimum set crossover probabilities is multiplied by the iteration rate to obtain a third product. The difference between the maximum set crossover probability and the third product is used as the crossover probability for the current iteration; otherwise, the maximum set crossover probability is used as the crossover probability for the current iteration. In each iteration, if the fitness value of the individual undergoing mutation is greater than the average fitness value of the population, the difference between the maximum and minimum set mutation probabilities is multiplied by the iteration rate to obtain the fourth product. The sum of the minimum set mutation rate and the third product is used as the mutation probability of the current iteration; otherwise, the minimum set mutation probability is used as the mutation probability of the current iteration. The iteration rate is equal to the current iteration round divided by the total number of iteration rounds.

[0016] This invention also proposes a hybrid AC / DC grid partitioning system considering modularity and reactive power margin, including an initial partitioning scheme generation module, an optimized partitioning scheme generation module, and a final partitioning scheme acquisition module: The initial partitioning scheme generation module extracts nodes in the AC / DC hybrid power grid and constructs the adjacency matrix of the AC / DC hybrid power grid; based on the connectivity of the nodes in the adjacency matrix, it generates several initial partitioning schemes. The partition scheme generation module optimizes the overall weight index of the initial partition scheme as the fitness value; based on the fitness value, several optimized partition schemes are selected. The final partitioning scheme acquisition module assigns corresponding crossover and mutation probabilities based on the fitness value of the optimized partitioning scheme; it iterates continuously until the maximum number of iterations is reached to obtain the final partitioning scheme. The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is used to perform the steps of the above method according to the instructions.

[0017] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention introduces modularity and reactive power regulation margin to form a comprehensive weight index, considering regional structural strength and the rational distribution of reactive power regulation resources, to obtain an optimized zoning scheme for AC / DC hybrid power grids. The modularity index in the proposed method describes the topology and node coupling relationship of the AC / DC hybrid power grid, while the regional voltage regulation capability index mainly describes the response time and regulation capacity of reactive power regulation equipment. Based on the principle of tight coupling within zoning and sparse coupling between regions, and taking into account regional autonomy of reactive power, the comprehensive weight of the index is determined using order relation analysis and inverse entropy weighting. Finally, an improved genetic algorithm is used for solving the problem. The resulting zoning can effectively improve the voltage level of AC / DC hybrid power grids, reduce the cross-regional active and reactive power transmission capacity, and provide theoretical support for power grid reactive power optimization and voltage control. Attached Figure Description

[0019] Figure 1 This is a flowchart of the AC / DC hybrid power grid partitioning method of the present invention, which considers modularity and reactive power margin. Detailed Implementation

[0020] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of this invention. The embodiments described in this application are merely some embodiments of this invention, and not all embodiments. Based on the spirit of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the protection scope of this invention.

[0021] The present invention proposes the following solution: like Figure 1 As shown, this invention proposes a zoning method for AC / DC hybrid power grids that considers modularity and reactive power margin, specifically including: Extract the nodes in the AC / DC hybrid power grid and construct the adjacency matrix of the AC / DC hybrid power grid; based on the connectivity of the nodes in the adjacency matrix, generate several initial partitioning schemes; A comprehensive weight index is calculated for the initial partitioning scheme, which is used as the fitness value; based on the fitness value, several optimized partitioning schemes are selected. Based on the fitness value of the optimized partitioning scheme, assign corresponding crossover and mutation probabilities; iterate continuously until the maximum number of iterations is reached to obtain the final partitioning scheme; A comprehensive weight index is calculated for the initial partitioning scheme, which is used as the fitness value; based on the fitness value, several optimized partitioning schemes are selected and subjected to crossover and mutation. Based on the fitness value of the optimized partitioning scheme, assign corresponding crossover and mutation probabilities; iterate continuously until convergence to obtain the final partitioning scheme; Set corresponding maximum and minimum values ​​for the crossover probability and the mutation probability, respectively; In each iteration, if the fitness value of the two individuals undergoing crossover is greater than the average fitness value of the population, the difference between the maximum and minimum set crossover probabilities is multiplied by the iteration rate to obtain a third product. The difference between the maximum set crossover probability and the third product is used as the crossover probability for the current iteration; otherwise, the maximum set crossover probability is used as the crossover probability for the current iteration. In each iteration, if the fitness value of the individual undergoing mutation is greater than the average fitness value of the population, the difference between the maximum and minimum set mutation probabilities is multiplied by the iteration rate to obtain the fourth product. The sum of the minimum set mutation rate and the third product is used as the mutation probability of the current iteration; otherwise, the minimum set mutation probability is used as the mutation probability of the current iteration. The iteration rate is equal to the current iteration round divided by the total number of iteration rounds.

[0022] The comprehensive weighting index is related to the modularity index and the regional voltage regulation capability index of the optimized zoning scheme; Modularity index and the average value of voltage regulation capability index for all regions Each is assigned a weight coefficient, and the comprehensive weight index γ is a modularity index. and the average value of voltage regulation capability index for all regions The weighted sum of the weights, where the sum of the weight coefficients is 1.

[0023] The modularity index The calculation method is as follows: Calculate the voltage change at node i corresponding to the unit value of reactive power change at node j, and obtain the voltage sensitivity matrix to reactive power. Using the element in the j-th row and j-th column of the sensitivity matrix as the numerator and the element in the ith row and j-th column as the denominator, the ratio is obtained. Taking the logarithm of the ratio gives the influence degree of node j on node i. Based on the influence of node j on node i, the electrical distance between node i and node j is calculated; the electrical distance between node i and node j is then converted into inter-node edge weights. The modularity index is calculated using the edge weights between the nodes. .

[0024] The calculation methods for the electrical distance and the edge weights between nodes are as follows: In a hybrid AC / DC power grid, the influence of any node on node i is subtracted from the influence of any node on node j, resulting in n differences. The electrical distance between node i and node j is obtained by squaring all the differences, summing them, and then taking the square root. To satisfy the relationship between edge weights and electrical distances between nodes, i.e., the smaller the electrical distance, the larger the edge weight, for the electrical distance between nodes i and j, the normalized electrical distance is subtracted from 1 to obtain the edge weights between nodes; Calculate the difference between the edge weights between nodes and the expected edge weights; sum all the differences and divide by twice the total edge weights to obtain the modularity index. .

[0025] The modularity index is related to the electrical distance between different nodes in the adjacency matrix; the regional voltage regulation capability index is the average of the regional voltage regulation capability index theoretically achieved by all partitions in the optimized partitioning scheme; For a given region, its regional voltage regulation capability index is related to the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in that region and the voltage regulation capability of the active power of the photovoltaic unit. The voltage regulation capability of photovoltaic power generation or reactive power compensation equipment in the zone is summed with the voltage regulation capability of active power of photovoltaic units to obtain a summation result; the summation result is compared with 1, and the smaller one is taken as the regional voltage regulation capability index of the zone.

[0026] The calculation method for the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the aforementioned zone is as follows: compute nodes The reactive power compensation capacity of adjustable resources and nodes Unit value of reactive power change corresponding to node The first product of the voltage changes is summed with the first products corresponding to all nodes in the partition to obtain a first threshold; when the node with the largest voltage deviation in the partition... If the voltage deviation is less than or equal to the first threshold, then the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition is equal to 1; otherwise, the first threshold is used as the numerator, and the node with the largest voltage deviation in the partition is selected. The voltage limit is used as the denominator to obtain the first ratio. The first ratio of all nodes in the partition is summed to obtain the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition.

[0027] To ensure that each zone has the theoretical capability to regulate the regional voltage, the output sequence and regulation range of the reactive power regulation equipment within the zone are set.

[0028] The specific steps for setting the output sequence and adjustment range of the reactive power regulating equipment within the specified zone are as follows: Optimize the output of reactive power regulation equipment based on the adjustment response time, prioritizing the adjustment of dynamic reactive power photovoltaic power generation and SVG, and then adjusting static reactive power capacitor banks; The intersection of capacity and power factor constraints is used as the reactive power regulation range for photovoltaic power generation. The square root of the difference between the photovoltaic power generation capacity and the active power of photovoltaic power generation at node i is taken as the upper limit threshold of the reactive power of photovoltaic power generation at node i; a negative sign is added to the square root result as the lower limit threshold of the reactive power of photovoltaic power generation at node i. The power factor of distributed photovoltaic power is between the lower limit of the power factor and 1; The sum of the reactive power adjustment margins and the sum of the reactive power adjustment margins of SVG satisfy the following: The difference between the adjustable upper limit and the real-time output of the i-th SVG reactive power is calculated, and the difference results of all SVGs are summed. The sum is used as the total upper limit adjustment margin of SVG reactive power. The difference between the real-time output and the adjustable lower limit of the i-th SVG reactive power is calculated, and the difference results of all SVGs are summed. The sum is used as the total lower limit adjustment margin of SVG reactive power. The connected capacity of the capacitor bank at node i is between the upper and lower limits of the connected capacity. The optimal operating point of the i-th SVG satisfies the following condition: at the optimal operating point, the output value of the i-th SVG is the average of the adjustable upper limit and the adjustable lower limit of the reactive power of the i-th SVG.

[0029] This invention also proposes a hybrid AC / DC grid partitioning system considering modularity and reactive power margin, including an initial partitioning scheme generation module, an optimized partitioning scheme generation module, and a final partitioning scheme acquisition module: The initial partitioning scheme generation module extracts nodes in the AC / DC hybrid power grid and constructs the adjacency matrix of the AC / DC hybrid power grid; based on the connectivity of the nodes in the adjacency matrix, it generates several initial partitioning schemes. The partition scheme generation module optimizes the overall weight index of the initial partition scheme as the fitness value; based on the fitness value, several optimized partition schemes are selected. The final partitioning scheme acquisition module assigns corresponding crossover and mutation probabilities based on the fitness value of the optimized partitioning scheme; it iterates continuously until the maximum number of iterations is reached to obtain the final partitioning scheme. The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is used to perform the steps of the above method according to the instructions.

[0030] The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0031] Example 1 This invention proposes a method for partitioning AC / DC hybrid power grids that considers regional modularity and reactive power regulation margin.

[0032] Specifically, this method is structured based on the principle of tight coupling within partitions and sparse coupling between regions, and introduces a modularity index to describe the topology and node coupling relationship of AC / DC hybrid power grids. Unlike other region partitioning methods, this invention uses a modularity index to measure the community structure strength of complex networks, automatically generating the optimal number of partitions without requiring pre-setting. Modularity Index The definition is as follows: (1) In the formula: This represents the weight (edge ​​weight) of the edge connecting node i and node j. When node i and node j are directly connected... When not connected In this method, it is set as electrical distance; This represents the sum of the weights of all edges connected to node i; This represents the sum of the weights of all edges in the network. If node i and node j are in the same region, then the function... ,otherwise .

[0033] In this method, the edge weights of the AC / DC hybrid power grid are primarily determined by the electrical distance between nodes. To describe the tightness of electrical coupling between two nodes in the network, electrical distance is used as a key indicator, obtained through the sensitivity of voltage to reactive power. (2) In the formula: This is the voltage sensitivity matrix to reactive power; and These represent the voltage amplitude and reactive power change, respectively. Matrix The element in the i-th row and j-th column This represents the change in voltage at node i corresponding to the unit value of the change in reactive power at node j.

[0034] The influence of node j on node i between two nodes for: (3) This formula represents the ratio of the voltage change at node j to the voltage change at node i when the reactive power changes. The larger the value, the smaller the influence of node j on node i, meaning the greater the distance between the two nodes.

[0035] Considering that the relationship between two nodes depends not only on themselves but also on other nodes in the network, let the network have n nodes, and define the electrical distance between node i and node j. for: (4) Using the electrical distance between nodes as the edge weight to describe the modularity index can reflect not only the structural performance within a region but also the degree of electrical coupling between nodes within that region. To satisfy the relationship between the edge weight between nodes and the electrical distance—that is, the smaller the electrical distance, the larger the edge weight—the edge weight between nodes is set as follows: ,in for The maximum value.

[0036] To address voltage control issues, the region should possess a certain voltage regulation capability, while also considering regional autonomy in reactive power regulation. The regional voltage regulation capability index is introduced to primarily describe the response time and regulation capacity of reactive power regulation equipment: When the penetration rate of renewable energy output in a hybrid AC / DC power grid is too high, leading to voltage over-limit, the reactive power compensation capacity within the region should meet the needs of local reactive power balance as much as possible, and internal resources should be used as much as possible to smooth voltage fluctuations within the region and reduce the transmission of reactive power across regions. Regarding the issue of active power matching, the region's self-absorption capacity should be fully utilized to reduce the transmission of active power from the region to other regions.

[0037] Regional voltage regulation capability reflects the ability of various regulation resources within a region to cope with nodal voltage exceedances at a given time scale. In this method, it is assessed through the ability of reactive power compensation equipment and photovoltaic units to resolve regional voltage exceedances. (The last sentence appears to be incomplete and possibly refers to a separate topic: "Regional...") For example, regional voltage regulation capability index Defined as: (5) In the formula, Indicates the area Voltage regulation capability; For the region The voltage regulation capability of photovoltaic power generation or reactive power compensation equipment in the system; For the region Voltage regulation capability of active power of photovoltaic units.

[0038] (6) (7) In the formula, In the region The node with the largest voltage deviation The voltage exceeds the limit, but no voltage over-limit occurs. ; and Representing nodes respectively The reactive power compensation capacity and maximum active power capacity of the adjustable resources. For nodes in the voltage-to-reactive power sensitivity matrix Unit value of reactive power change corresponding to node The change in voltage; For nodes in the voltage-to-active power sensitivity matrix Unit value of active power change corresponding to node The change in voltage.

[0039] Average voltage regulation capability index across all regions for: (8) In the formula: N is the number of regions.

[0040] The reactive power regulation equipment within the region includes capacitor banks, photovoltaic power generation, and static var generators (SVG). The reactive power compensation target within the region is to fully utilize various reactive power regulation equipment under normal conditions to ensure that the system voltage does not exceed limits. Simultaneously, in the event of system disturbances or insufficient dynamic reactive power, the dynamic reactive power support capability of SVG (Static Var Generator) will be used to improve node voltage levels.

[0041] First, optimize the output of reactive power regulation equipment based on the adjustment response time. In principle, prioritize the adjustment of dynamic reactive power photovoltaic power generation and SVG, and then adjust the static reactive power capacitor bank. While reducing the number of capacitor bank switching, maximize the use of reactive power regulation equipment in the area to effectively alleviate node voltage fluctuations caused by load changes and other disturbances.

[0042] The reactive power capacity of photovoltaic (PV) power generation is limited by both capacity and power factor. The actual reactive power regulation range of PV power generation is the intersection of these two constraints. (9) In the formula: Let i be the capacity of photovoltaic power generation at network node i; Let be the active power of photovoltaic power generation at network node i; Let i be the reactive power generated by the photovoltaic power generation at network node i; The lower limit of the power factor for distributed photovoltaic systems is determined by the physical characteristics of the equipment and is a fixed value.

[0043] SVG reactive power upscaling margin total and the total margin reduction for: (10) In the formula: , , , respectively, represent the real-time reactive power output, adjustable upper limit, and adjustable lower limit of the i-th SVG; m is the number of SVGs.

[0044] Capacitor banks must meet their capacity constraints to operate: (11) In the formula: , , These represent the connected capacity of the capacitor bank at node i and its upper and lower limits, which are determined by the inherent properties of the equipment at the factory.

[0045] Further optimize the capacity allocation of reactive power regulation equipment to achieve synergy between dynamic reactive power (Static Var Generator, SVG) and static reactive power (capacitor bank) regulation resources, as well as fast reactive power (SVG) and slow reactive power (photovoltaic power generation) resources. This will allow the system to reserve a certain reactive power regulation margin, so that SVG and other equipment can quickly provide reactive power support to the system in the event of system failure or load change, and minimize the occurrence of node voltage exceeding limits.

[0046] The optimal running point of the i-th SVG is : (12) To optimize reactive power configuration, sufficient reserve capacity should be reserved for SVG (Static Var Generator) to cope with system failures or sudden load changes. This patent considers prioritizing the provision of basic reactive power by static reactive power regulation resources such as fixed switching capacitor banks, with the remainder supplemented by dynamic reactive power regulation resources such as SVG, to ensure that the system has a certain reactive power regulation margin. That is, when the SVG output is higher than the optimal operating point, the reactive power is prioritized to be provided by capacitor banks, reducing the SVG output and allowing the SVG to operate at the optimal operating point, thus retaining a certain reactive power regulation margin.

[0047] The weighted average of the modularity index and the regional voltage regulation capability index forms the following comprehensive index: This patent proposes a regional voltage regulation capability index that takes into account both regional structural modularity and reactive power regulation capability. The comprehensive weighting index γ proposed in this method is: (13) In the formula: and The weighting coefficient for the indicator.

[0048] The combined weights of the indicators were determined using order relation analysis and inverse entropy weighting as follows: The main steps for determining the subjective weights of indicators using ordinal relation analysis are as follows.

[0049] Step 1: Experts rank the indicators by importance (e) according to the evaluation criteria for the zoning indicators. (14) In the formula, e represents the expert's evaluation of the importance of the indicator.

[0050] Step 2: The ratio of the importance of two adjacent indicators Assign a value. When the indicator With indicators Equally important, take When the indicator Comparison Indicators Slightly important, take When the indicator Comparison Indicators Clearly important, take When the indicator Comparison Indicators Strongly important, take When the indicator Comparison Indicators Extremely important, take .

[0051] Step 3: Calculate the subjective weight of the indicator with the lowest relative importance. for: (15) Depend on The subjective weights of other relatively important indicators, derived recursively, are as follows: (16) The main steps for determining the objective weights of indicators using the inverse entropy weight method are as follows.

[0052] Step 1: Calculate the weight of each indicator within its category of indicators. for: (17) (18) In the formula: Let be the original value of the i-th partition scheme on the j-th indicator, i=1,2,…,m, where m is the number of evaluation objects, and j=1,2,…,n; Let be the standardized value of the i-th partition scheme on the j-th indicator.

[0053] Step 2: Calculate the inverse entropy value of the j-th index. for: (19) like Then define .

[0054] Step 3: Calculate the objective weight of the j-th indicator. for: (20) Based on the principle of minimum entropy weighting, and by weighting the above subjective and objective weights, the combined weights of the zoning scheme indicators are obtained as follows: (twenty one) In the formula: The combined weight of the j-th indicator; Let be the subjective weight of the j-th indicator; Let be the objective weight of the j-th indicator.

[0055] The partitioning results of the AC / DC hybrid power grid obtained based on the improved genetic algorithm are as follows: Compared to conventional partitioning algorithms, the global search capability of genetic algorithms ensures that they gradually approach the global optimum with increasing iterations. When applied to dynamic region partitioning, the partitioning comprehensive index is used as the fitness function, and the region partitioning result is treated as the problem to be solved for optimization. The solution of the genetic algorithm is the final result of the region partitioning.

[0056] Considering that the regional division problem cannot change the original network structure of the AC / DC hybrid power grid, i.e., the connectivity constraints of nodes within the region, this method encodes chromosomes based on the network's adjacency matrix. The network's adjacency matrix represents the connectivity of nodes in the network, containing only 0 and 1 elements, where 0 indicates no connection between nodes and 1 indicates connection between nodes. This encoding method significantly reduces the search range of the genetic algorithm and lowers the search time while ensuring node connectivity.

[0057] To improve the convergence speed and global search capability of the genetic algorithm, the crossover and mutation rates are adaptively adjusted. Specifically, if an individual's fitness value is less than the average fitness value, it is given a higher crossover probability and a lower mutation probability; if an individual's fitness value is greater than the average fitness value, it is assigned a corresponding crossover and mutation probability based on its iteration state. Considering that as the number of iterations increases, the partitioning results of different regions become more similar, excessively high crossover probabilities become meaningless. Therefore, the mutation probability should be appropriately increased to enhance the algorithm's local search capability. The crossover and mutation probabilities used are: (twenty two) (twenty three) In the formula: and These represent the crossover and mutation probabilities, respectively. , and , These represent the maximum and minimum probabilities of crossover and mutation, respectively. This represents the number of iterations. This represents the maximum number of iterations. The larger fitness value between the two individuals undergoing the crossover operation; The fitness value of the individual to be mutated; This represents the average fitness of the population.

[0058] Example 2 This invention also proposes a hybrid AC / DC grid partitioning system considering modularity and reactive power margin, including an initial partitioning scheme generation module, an optimized partitioning scheme generation module, and a final partitioning scheme acquisition module: The initial partitioning scheme generation module extracts nodes in the AC / DC hybrid power grid and constructs the adjacency matrix of the AC / DC hybrid power grid; based on the connectivity of the nodes in the adjacency matrix, it generates several initial partitioning schemes. The partition scheme generation module optimizes the overall weight index of the initial partition scheme as the fitness value; based on the fitness value, several optimized partition schemes are selected. The final partitioning scheme acquisition module assigns corresponding crossover and mutation probabilities based on the fitness value of the optimized partitioning scheme; it iterates continuously until the maximum number of iterations is reached to obtain the final partitioning scheme. Example 3 The present invention also proposes a terminal, including a processor and a storage medium: The storage medium is used to store instructions; The processor is used to perform the steps of the above method according to the instructions.

[0059] Example 4 The present invention also proposes a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the above-described method.

[0060] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the specific implementation of the present invention. Any modifications or equivalent substitutions that do not depart from the spirit and scope of the present invention should be covered within the protection scope of the claims of the present invention.

Claims

1. A partitioning method for AC / DC hybrid power grids considering modularity and reactive power margin, characterized in that, include: Extract the nodes in the AC / DC hybrid power grid and construct the adjacency matrix of the AC / DC hybrid power grid; Based on the connectivity of nodes in the adjacency matrix, several initial partitioning schemes are generated; A comprehensive weight index is calculated for the initial partitioning scheme as a fitness value; based on the fitness value, several optimized partitioning schemes are selected; wherein, the comprehensive weight index includes a modularity index and a regional voltage regulation capability index; Based on the fitness value of the optimized partitioning scheme, assign corresponding crossover and mutation probabilities; iterate continuously until the maximum number of iterations is reached to obtain the final partitioning scheme; The modularity index describes the coupling relationship between the AC / DC hybrid power grid topology and nodes; the regional voltage regulation capability index characterizes the regional voltage regulation capability. When calculating the voltage regulation capability index of the region, the output sequence of the reactive power regulation equipment and the regulation range of the reactive power regulation equipment are set within the zone.

2. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 1, characterized in that: Modularity index and the average value of voltage regulation capability index for all regions Each is assigned a weight coefficient, and the comprehensive weight index γ is a modularity index. and the average value of voltage regulation capability index for all regions The weighted sum of the weights, where the sum of the weight coefficients is 1.

3. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 1, characterized in that: The modularity index The calculation method is as follows: Calculate the voltage change at node i corresponding to the unit value of reactive power change at node j, and obtain the voltage sensitivity matrix to reactive power. Using the element in the j-th row and j-th column of the sensitivity matrix as the numerator and the element in the ith row and j-th column as the denominator, the ratio is obtained. Taking the logarithm of the ratio gives the influence degree of node j on node i. Based on the influence of node j on node i, the electrical distance between node i and node j is calculated; the electrical distance between node i and node j is then converted into inter-node edge weights. The modularity index is calculated using the edge weights between the nodes. .

4. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 3, characterized in that: The calculation methods for the electrical distance and the edge weights between nodes are as follows: In a hybrid AC / DC power grid, the influence of any node on node i is subtracted from the influence of any node on node j, resulting in n differences. The electrical distance between node i and node j is obtained by squaring all the differences, summing them, and then taking the square root. To satisfy the relationship between edge weights and electrical distances between nodes, i.e., the smaller the electrical distance, the larger the edge weight, for the electrical distance between nodes i and j, the normalized electrical distance is subtracted from 1 to obtain the edge weights between nodes; Calculate the difference between the edge weights between nodes and the expected edge weights; sum all the differences and divide by twice the total edge weights to obtain the modularity index. .

5. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 1, characterized in that: For a given region, its regional voltage regulation capability index is related to the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in that region and the voltage regulation capability of the active power of the photovoltaic unit. The voltage regulation capability of photovoltaic power generation or reactive power compensation equipment in the zone is summed with the voltage regulation capability of active power of photovoltaic units to obtain a summation result; the summation result is compared with 1, and the smaller one is taken as the regional voltage regulation capability index of the zone.

6. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 5, characterized in that: The calculation method for the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the aforementioned zone is as follows: compute nodes The reactive power compensation capacity of adjustable resources and nodes Unit value of reactive power change corresponding to node The first product of the voltage change values ​​is summed with the first products corresponding to all nodes in the partition to obtain the first threshold. When the node with the largest voltage deviation in the partition If the voltage limit is not greater than the first threshold, then the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition is equal to 1. Conversely, using the first threshold as the numerator, the node with the largest voltage deviation in the partition is selected. The voltage limit is used as the denominator to obtain the first ratio. The first ratio of all nodes in the partition is summed to obtain the voltage regulation capability of the photovoltaic power generation or reactive power compensation equipment in the partition.

7. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 5, characterized in that: The calculation method for the voltage regulation capability of the active power of the photovoltaic units in the aforementioned zone is as follows: compute nodes The maximum active power capacity of adjustable resources and nodes Unit value of active power change corresponding to node The second product of the voltage change values ​​is summed by adding the second products corresponding to all nodes in the partition to obtain the second threshold. When the node with the largest voltage deviation in the partition If the voltage limit is not greater than the second threshold, then the voltage regulation capability of the active power of the photovoltaic units in the partition is equal to 1. Conversely, using the second threshold as the numerator, the node with the largest voltage deviation in the partition is selected. The voltage limit is used as the denominator to obtain the second ratio. The second ratios of all nodes in the partition are summed to obtain the voltage regulation capability of the active power of the photovoltaic unit in the partition.

8. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 1, characterized in that: The specific steps for setting the output sequence and adjustment range of the reactive power regulating equipment within the specified zone are as follows: Optimize the output of reactive power regulation equipment based on the adjustment response time, prioritizing the adjustment of dynamic reactive power photovoltaic power generation and SVG, and then adjusting static reactive power capacitor banks; The intersection of capacity and power factor constraints is used as the reactive power regulation range for photovoltaic power generation. The square root of the difference between the photovoltaic power generation capacity and the active power of photovoltaic power generation at node i is taken as the upper limit threshold of the reactive power of photovoltaic power generation at node i; a negative sign is added to the square root result as the lower limit threshold of the reactive power of photovoltaic power generation at node i. The power factor of distributed photovoltaic power is between the lower limit of the power factor and 1; The sum of the reactive power adjustment margins and the sum of the reactive power adjustment margins of SVG satisfy the following: The difference between the adjustable upper limit and the real-time output of the i-th SVG reactive power is calculated, and the difference results of all SVGs are summed. The sum is used as the total upper limit adjustment margin of SVG reactive power. The difference between the real-time output and the adjustable lower limit of the i-th SVG reactive power is calculated, and the difference results of all SVGs are summed. The sum is used as the total lower limit adjustment margin of SVG reactive power. The connected capacity of the capacitor bank at node i is between the upper and lower limits of the connected capacity. The optimal operating point of the i-th SVG satisfies the following condition: at the optimal operating point, the output value of the i-th SVG is the average of the adjustable upper limit and the adjustable lower limit of the reactive power of the i-th SVG.

9. The AC / DC hybrid power grid partitioning method considering modularity and reactive power margin according to claim 1, characterized in that: A comprehensive weight index is calculated for the initial partitioning scheme, which is used as the fitness value; based on the fitness value, several optimized partitioning schemes are selected and subjected to crossover and mutation. Based on the fitness value of the optimized partitioning scheme, assign corresponding crossover and mutation probabilities; iterate continuously until convergence to obtain the final partitioning scheme; Set corresponding maximum and minimum values ​​for the crossover probability and the mutation probability, respectively; In each iteration, if the fitness value of the two individuals undergoing crossover is greater than the average fitness value of the population, the difference between the maximum and minimum set crossover probabilities is multiplied by the iteration rate to obtain a third product. The difference between the maximum set crossover probability and the third product is used as the crossover probability for the current iteration; otherwise, the maximum set crossover probability is used as the crossover probability for the current iteration. In each iteration, if the fitness value of the individual undergoing mutation is greater than the average fitness value of the population, the difference between the maximum and minimum set mutation probabilities is multiplied by the iteration rate to obtain the fourth product. The sum of the minimum set mutation rate and the third product is used as the mutation probability of the current iteration; otherwise, the minimum set mutation probability is used as the mutation probability of the current iteration. The iteration rate is equal to the current iteration round divided by the total number of iteration rounds.

10. A hybrid AC / DC grid partitioning system considering modularity and reactive power margin using the method described in any one of claims 1-9, comprising an initial partitioning scheme generation module, an optimized partitioning scheme generation module, and a final partitioning scheme acquisition module, characterized in that: The initial partitioning scheme generation module extracts nodes in the AC / DC hybrid power grid and constructs the adjacency matrix of the AC / DC hybrid power grid; based on the connectivity of the nodes in the adjacency matrix, it generates several initial partitioning schemes. The partition scheme generation module optimizes the overall weight index of the initial partition scheme as the fitness value; based on the fitness value, several optimized partition schemes are selected. The final partitioning scheme acquisition module assigns corresponding crossover and mutation probabilities based on the fitness value of the optimized partitioning scheme; it iterates continuously until the maximum number of iterations is reached to obtain the final partitioning scheme.

11. A terminal, comprising a processor and a storage medium; characterized in that: The storage medium is used to store instructions; The processor is configured to operate according to the instructions to perform the steps of the method according to any one of claims 1-9.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by a processor, the program implements the steps of the method according to any one of claims 1-9.

Images