Node partitioning method, system, device and medium for ac-dc hybrid distribution network

Abstract

The application discloses a node partitioning method, system, device and medium for an AC-DC hybrid power distribution network, and is characterized in that the method comprises the following steps: acquiring AC line parameters, DC line parameters and converter operation data of the AC-DC hybrid power distribution network; calculating AC electrical distances according to the AC line parameters and calculating DC electrical distances according to the DC line parameters; determining a converter control constraint equation based on the converter operation data, so that the converter control constraint equation and a power balance equation form a power flow equation, performing power flow calculation on the power flow equation, determining sensitivity distances between node pairs, calculating edge weights between node pairs based on the AC electrical distances, the DC electrical distances and the sensitivity distances, and iteratively optimizing each initial node partitioning scheme until a node partitioning scheme with the largest fitness value is output as a partitioning result of the AC-DC hybrid power distribution network when a convergence condition is met. The application can improve the accuracy of the partitioning result of the AC-DC hybrid power distribution network.

Description

Technical Field

[0001] This invention relates to the field of distribution network cluster partitioning, and more particularly to node partitioning methods, systems, equipment and media for AC / DC hybrid distribution networks. Background Technology

[0002] With the large-scale integration of distributed power sources, energy storage devices, and flexible loads, distribution networks are gradually evolving from traditional AC distribution networks to hybrid AC / DC distribution networks. In hybrid AC / DC distribution networks, AC lines, DC lines, and AC / DC connection equipment all participate in power transmission and regulation, making the network structure and operating characteristics more complex. To improve the reliability, fault isolation capability, and distributed resource coordination and control capability of the distribution network, it is usually necessary to partition the distribution network according to the electrical interconnection characteristics between nodes, grouping electrically connected nodes into the same area, thereby reducing power exchange between areas and improving the autonomous operation capability of each area.

[0003] However, most existing partitioning methods are based on AC network models, making it difficult to uniformly incorporate actual operating DC lines into the node correlation analysis process. Some methods directly ignore the influence of DC lines, while others simply treat DC lines as equivalent to AC branches, resulting in the power controllability and cross-regional power regulation capabilities of DC lines not being reflected in the partitioning results. In addition, the control methods of AC and DC connection equipment directly affect the node voltage distribution, power flow direction, and the degree of mutual influence between nodes. When establishing node correlation models, existing partitioning methods usually do not fully consider the impact of converter control methods on network operating status, resulting in deviations between the obtained node correlation characteristics and the actual operating status, thus affecting the rationality of the final partitioning results. Summary of the Invention

[0004] This invention provides a node partitioning method, system, equipment, and medium for AC / DC hybrid distribution networks. It can combine the influence of DC line characteristics and converter control methods on node correlation characteristics, thereby improving the accuracy of AC / DC hybrid distribution network partitioning results.

[0005] In a first aspect, embodiments of the present invention provide a node partitioning method for an AC / DC hybrid distribution network, comprising: Acquire AC line parameters, DC line parameters, and converter operating data of the AC / DC hybrid distribution network; Calculate the AC electrical distance between each AC node pair and each node pair at the AC / DC interface based on the AC line parameters; calculate the DC electrical distance between each DC node pair based on the DC line parameters. Based on the converter operating data, the converter control constraint equations are determined. These equations are then combined with preset AC node power balance equations and DC node power balance equations to form a unified AC / DC power flow equation. Power flow calculations are performed on this unified AC / DC power flow equation to determine the sensitivity distance between each pair of nodes in the AC / DC hybrid distribution network. The edge weights between each pair of nodes are calculated based on the AC electrical distance, DC electrical distance, and sensitivity distance. The fitness values ​​corresponding to each initial node partitioning scheme are calculated based on the edge weights between each pair of nodes. The initial node partitioning schemes are iteratively optimized using a genetic algorithm until a preset convergence condition is met. The node partitioning scheme with the largest fitness value is then output as the partitioning result of the AC / DC hybrid distribution network. Each initial node partitioning scheme is determined based on the network topology of the AC / DC hybrid distribution network.

[0006] This invention provides a complete data foundation for subsequent multi-type line modeling and control characteristic analysis by acquiring structural parameters and operational control information in AC / DC hybrid distribution networks. This ensures the integrity of the input for zoning analysis from the source, thereby improving the accuracy of the zoning results for AC / DC hybrid distribution networks. By converting AC and DC lines into quantifiable inter-node electrical correlation strengths, lines with different electrical characteristics can be expressed within the same analytical framework, further improving the accuracy of the zoning results for AC / DC hybrid distribution networks. Introducing converter control methods into the power flow calculation model allows the operational constraints of the AC / DC system and its control equipment to be considered simultaneously within a unified power flow framework, thus improving the accuracy of the zoning results for AC / DC hybrid distribution networks. The power flow calculation results reflect the inter-node... The power disturbance response relationship allows node correlations to reflect not only structural characteristics but also the impact of operational status, thereby improving the accuracy of AC / DC hybrid distribution network zoning results. By fusing structural distance and operational sensitivity information, a unified quantitative expression of node correlations is provided, enhancing the comprehensiveness and consistency of node correlation characterization, thus improving the accuracy of AC / DC hybrid distribution network zoning results. By transforming the continuous node correlation problem into a discrete, optimizable combinatorial representation, an initial solution set is provided for subsequent global optimization searches, thereby improving the accuracy of AC / DC hybrid distribution network zoning results. Through a global iterative optimization mechanism, the optimal solution is searched among multiple candidate zoning schemes, enabling the zoning results to reach the optimal level under comprehensive evaluation indicators, thus improving the accuracy of AC / DC hybrid distribution network zoning results.

[0007] Furthermore, the calculation of the AC electrical distance between each AC node pair and between each node pair at the AC / DC interface based on the AC line parameters includes: Based on the AC line parameters, a node admittance matrix is ​​constructed to characterize the line connection relationship between nodes in the AC network. The node impedance matrix is ​​obtained by inverting the node admittance matrix. Extract the self-impedance parameters of each AC node and each AC / DC interface node, as well as the mutual impedance parameters between any pair of nodes, from the node impedance matrix. Based on the self-impedance parameters and mutual impedance parameters, the AC electrical distance between each AC node pair and the node pairs at the AC / DC interface is calculated.

[0008] This invention constructs an AC electrical distance, transforming the electrical correlation between nodes from a line parameter expression into a unified electrical distance metric, thereby more accurately reflecting the intensity of electrical influence between nodes and improving the accuracy of AC / DC hybrid distribution network zoning results.

[0009] Furthermore, the calculation of the DC electrical distance between each pair of DC nodes based on the DC line parameters includes: Obtain the line resistance, rated voltage, and rated capacity of each DC line, and calculate the equivalent transmission current of each DC line based on the rated voltage and rated capacity. Calculate the equivalent voltage drop on each DC line based on the equivalent transmission current and the line resistance described above. Based on the equivalent voltage drop and the preset DC network topology path information, the DC electrical distance between each pair of DC nodes is calculated.

[0010] The embodiments of the present invention express the operational constraints and electrical effects of DC networks in a unified quantitative form, thereby more accurately reflecting the actual electrical impedance and power transmission characteristics between DC nodes, and thus improving the accuracy of the zoning results of AC / DC hybrid distribution networks.

[0011] Furthermore, the step of performing power flow calculations on the unified AC / DC power flow equations to determine the sensitivity distance between node pairs in the AC / DC hybrid distribution network includes: The power flow calculation of the AC / DC unified power flow equation is performed using the Newton-Raphson iteration method to obtain the unified state variable convergent solution of the AC / DC hybrid distribution network, and a unified Jacobian matrix is ​​constructed based on the unified state variable convergent solution. The unified Jacobian matrix is ​​inverted to obtain a sensitivity matrix including the state variables of AC nodes, DC nodes, and AC / DC interface nodes. The sensitivity distance between each node pair in the AC / DC hybrid distribution network is calculated based on the sensitivity matrix.

[0012] This invention transforms the description of the electrical response relationship between nodes from static parameters to a dynamic disturbance response quantification, thereby accurately characterizing the voltage-reactive power coupling characteristics between nodes. As a result, it can more accurately reflect the differences in electrical sensitivity between nodes in AC / DC hybrid distribution networks, thus improving the accuracy of AC / DC hybrid distribution network zoning results.

[0013] Furthermore, the calculation of the edge weights between each pair of nodes based on the AC electrical distance, the DC electrical distance, and the sensitivity distance includes: The AC electrical distance, DC electrical distance, and sensitivity distance are weighted and summed to obtain a unified electrical distance value for each node pair. The uniform electrical distance values ​​are superimposed with a preset non-zero regularization constant to obtain the distance correction parameters for each node pair. The reciprocal of the distance correction parameter is used to obtain the edge weights between each pair of nodes, which characterize the electrical coupling strength between nodes.

[0014] This invention converts distance relationships into edge weights, thereby achieving a stable mapping from multi-source electrical characteristics to a unified coupling strength. This makes the electrical correlation between nodes more comprehensive and comparable, thus improving the accuracy of AC / DC hybrid distribution network zoning results.

[0015] Furthermore, the step of iteratively optimizing each of the initial node partitioning schemes using a genetic algorithm until a preset convergence condition is met, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network, includes: The adjacency matrix encoding corresponding to each initial node partitioning scheme is parsed to determine the cluster to which each node belongs in the AC / DC hybrid distribution network. Based on the edge weights between each cluster and each node pair, the weighted modularity index corresponding to each initial node partitioning scheme is calculated. Based on the power operation data of each cluster and each node, calculate the source-load balance index corresponding to each initial node partitioning scheme. Based on the external power exchange data of each cluster and each node, the interface fluctuation index corresponding to each initial node partitioning scheme is calculated, wherein the power operation data and the external power exchange data are calculated based on the converter operation data. The weighted modularity index, the source-load balance index, and the interface fluctuation index are weighted and summed according to preset weight coefficients to obtain the fitness value corresponding to each initial node partitioning scheme. A genetic algorithm is used to iteratively optimize the node partitioning schemes corresponding to each fitness value until the preset convergence condition is met. The node partitioning scheme with the largest fitness value is then output as the partitioning result of the AC / DC hybrid distribution network.

[0016] This invention describes the internal electrical coupling structure of a cluster by calculating a weighted modularity index based on the edge weight node partitioning scheme. It also calculates a source-load balance index to reflect the cluster's self-balancing capability by combining load and distributed power source operation data, and introduces an interface fluctuation index to characterize external interaction stability by introducing external net exchange power. Furthermore, it forms a fitness function by weighted fusion of multiple indices and uses a genetic algorithm for iterative optimization, thereby achieving global optimization of the partitioning scheme. Therefore, it can improve the accuracy of the partitioning results of AC / DC hybrid distribution networks.

[0017] Furthermore, the step of iteratively optimizing the node partitioning schemes corresponding to each fitness value using a genetic algorithm until a preset convergence condition is met, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network, includes: By selecting the partitioning scheme for each node, a set of parent individuals is obtained; The gene codes of the parent generation set are cross-processed using a simulated binary crossover operator to obtain the first generation set of offspring individuals; The gene loci of the first offspring set are perturbed using a polynomial mutation operator to obtain the second offspring set. The second set of offspring individuals is subjected to connectivity constraint verification until the verification result meets the preset convergence condition. Then, the node partitioning scheme with the largest fitness value is output as the partitioning result of the AC / DC hybrid distribution network.

[0018] This invention improves the accuracy of AC / DC hybrid distribution network partitioning by using a collaborative iterative optimization of partitioning schemes through selection, crossover, and mutation, and by introducing constraint corrections during the process to ensure that the partitioning results meet the operation and topology conditions of AC / DC hybrid distribution networks. Finally, it outputs the optimal fitness partitioning results.

[0019] Secondly, embodiments of the present invention provide a node partitioning system for an AC / DC hybrid distribution network, the node partitioning system being used to perform the node partitioning method for an AC / DC hybrid distribution network as described in this application.

[0020] Thirdly, embodiments of the present invention provide a terminal device, including: a processor, a memory, a communication interface, and a communication bus, wherein the processor, the memory, and the communication interface communicate with each other through the communication bus; The memory is used to store at least one executable instruction that causes the processor to perform operations of the node partitioning method for the AC / DC hybrid distribution network as described in this application.

[0021] Fourthly, embodiments of the present invention provide a computer-readable storage medium comprising a stored computer program, wherein, when the computer program is executed, it controls the device or system where the computer-readable storage medium is located to perform the node partitioning method for an AC / DC hybrid distribution network as described in this application.

[0022] The above description is merely an overview of the technical solutions of the embodiments of the present invention. In order to better understand the technical means of the embodiments of the present invention and to implement them in accordance with the contents of the specification, and to make the above and other objects, features and advantages of the embodiments of the present invention more apparent and understandable, specific embodiments of the present invention are described below. Attached Figure Description

[0023] 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.

[0024] Figure 1 This is a flowchart illustrating an embodiment of the node partitioning method for an AC / DC hybrid distribution network provided in this application; Figure 2 This is a flowchart illustrating steps S201 to S203 provided in this application; Figure 3 This is a flowchart illustrating steps S301 to S303 provided in this application; Figure 4 This is a flowchart illustrating steps S401 to S405 provided in this application. Detailed Implementation

[0025] 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.

[0026] 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.

[0027] 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.

[0028] 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.

[0029] 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.

[0030] 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).

[0031] 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.

[0032] With the large-scale integration of distributed power sources, energy storage, and flexible loads, distribution networks are gradually evolving into hybrid AC / DC structures. AC lines, DC lines, and AC / DC interface equipment such as converters jointly participate in power transmission and regulation, making the network structure and operating mechanism more complex. To improve operational reliability and distributed resource coordination capabilities, it is usually necessary to partition the network based on the electrical correlation characteristics of nodes. However, existing methods are mostly based on AC models, making it difficult to uniformly characterize the power controllability characteristics and cross-regional regulation effects of DC lines, or they only perform simple equivalent processing, resulting in a weakening of DC characteristics. At the same time, existing methods do not adequately consider the control methods of AC / DC interface equipment such as converters, failing to fully reflect their impact on voltage distribution and power flow. This leads to discrepancies between node correlation modeling and actual operating conditions, thereby reducing the rationality and accuracy of partitioning results.

[0033] See Figure 1 In order to improve the accuracy of AC / DC hybrid distribution network zoning results by combining the influence of DC line characteristics and converter control mode on node association characteristics, an embodiment of the present invention provides a node zoning method for AC / DC hybrid distribution networks, including steps S101 to S103. Step S101: Obtain AC line parameters, DC line parameters, and converter operation data of the AC / DC hybrid distribution network; In some embodiments, the topology data of the AC / DC hybrid distribution network to be analyzed is read, a correspondence between node numbers and branch numbers is established, and the branch type corresponding to each branch is identified to determine whether each branch belongs to an AC line, a DC line, or an AC / DC connection branch. Based on the identified AC lines, the corresponding line parameters are read, including line resistance, line reactance, and ground susceptance. For transformers in the AC network, the corresponding turns ratio and short-circuit impedance parameters are further read, and the line parameters are associated and stored according to the corresponding branch numbers to form an AC line parameter set. Based on the identified DC lines, the corresponding line parameters are read, including line resistance, rated voltage, line length, and allowable transmission power. The DC line connection relationship is established by combining the starting and ending nodes of each DC line to form a DC line parameter set. The equipment parameters corresponding to the AC / DC connection equipment are read, and the connection node information, rated capacity, loss parameters, and control mode identifier of each converter are obtained to determine the connection relationship between each converter and the AC and DC networks.

[0034] In some embodiments, based on the control mode identifier corresponding to each converter, the corresponding operating data is read. The converter operating data includes active power reference value, reactive power reference value, DC voltage reference value, and droop coefficient. When the converter operates in a combined control mode, the reference value parameters corresponding to each control mode are read respectively, and a set of operating data for the corresponding converter is formed. The acquired AC line parameters, DC line parameters, and converter operating data are uniformly converted to a preset per-unit value benchmark, and the consistency of node number, branch number, line parameters, and equipment parameters is verified to obtain standardized AC line parameters, standardized DC line parameters, and standardized converter operating data, which serve as the basic input data for subsequent AC electrical distance calculation, DC electrical distance calculation, and construction of converter control constraint equations.

[0035] Through the above steps, by uniformly acquiring, standardizing, and verifying the AC line parameters, DC line parameters, and converter operation data of the AC / DC hybrid distribution network, different types of network data can be structurally expressed and parameter aligned under the same per-unit system. This ensures the consistency and accuracy of the input data required for subsequent electrical distance calculation and control constraint modeling, and improves the reliability of subsequent analysis and zoning results of the AC / DC hybrid distribution network.

[0036] Step S102: Calculate the AC electrical distance between each AC node pair and each node pair at the AC / DC interface based on the AC line parameters; calculate the DC electrical distance between each DC node pair based on the DC line parameters. In some embodiments, calculating the AC electrical distance between each AC node pair and each node pair at the AC / DC interface based on the AC line parameters includes: constructing a node admittance matrix to characterize the line connection relationship between each node in the AC network based on the AC line parameters; performing an inversion operation on the node admittance matrix to obtain a node impedance matrix; extracting the self-impedance parameters corresponding to each AC node and each AC / DC interface node, as well as the mutual impedance parameters between any node pairs, from the node impedance matrix; and calculating the AC electrical distance between each AC node pair and each node pair at the AC / DC interface based on the self-impedance parameters and the mutual impedance parameters.

[0037] In some embodiments, a node admittance matrix is ​​constructed based on the AC line parameters to characterize the line connection relationships between nodes in the AC network. The node admittance matrix is ​​then inverted to obtain the node impedance matrix. Specifically, based on the AC lines existing in the distribution network, for any two nodes i and j, when there is a direct AC line connection between nodes i and j, the node impedance matrix is ​​calculated based on the resistance of that AC line. With reactance Construct the series impedance of this line: ,in, This represents the complex impedance of the AC line between node i and node j. This represents the resistance of the AC line between node i and node j. Let represent the reactance of the AC line between node i and node j, where j represents the imaginary unit; thus, the admittance of this line is: ,in, This represents the off-diagonal admittance element between node i and node j; when there is no direct AC line connection between node i and node j, the admittance value between this node pair is: For the diagonal elements of the node admittance matrix, for any node i, its diagonal elements are obtained by summing the admittances of all AC lines directly connected to that node, while also considering the admittance of the node's own ground branches. Its expression is as follows: ,in, Let i represent the set of all nodes that have a direct communication line connection with node i. This represents the equivalent ground admittance or parallel branch admittance at node i. This represents the self-admittance of node i. , These represent the resistance and reactance of the AC line, respectively. After calculating the admittance elements of all node pairs one by one using the above method, a node admittance matrix is ​​formed: Here, matrix Y represents the electrical connection strength relationship between nodes in the AC section of the distribution network. Performing matrix inversion on the node admittance matrix yields the node impedance matrix: The node impedance matrix is ​​used to characterize the coupling relationship between the injected current and the node voltage at any node in an AC network.

[0038] In some embodiments, extracting the self-impedance parameters of each AC node and each AC / DC interface node, as well as the mutual impedance parameters between any pair of nodes, from the node impedance matrix specifically involves: obtaining the node impedance matrix... Then, the matrix elements are directly analyzed: For any node i, its self-impedance parameter is defined as the diagonal element of the node impedance matrix: The self-impedance parameter characterizes the equivalent electrical response of a node in the entire AC network. For any different nodes i and j, their mutual impedance parameters are defined as the off-diagonal elements of the node impedance matrix: The mutual impedance parameter characterizes the electrical coupling relationship between node i and node j through the AC network. When the AC network impedance matrix satisfies the symmetry condition, we have: Thus, a complete set of self-impedance parameters and mutual impedance parameters are obtained.

[0039] In some embodiments, the AC electrical distance between each AC node pair and between each node pair at the AC / DC interface is calculated based on each self-impedance parameter and each mutual impedance parameter. Specifically, for any node pair... The AC electrical distance is calculated based on the elements of the node impedance matrix and expressed as: ,in, This represents the AC electrical distance between node i and node j. This represents the self-impedance parameter of node i. This represents the self-impedance parameter of node j. This represents the mutual impedance parameter between node i and node j. Let represent the mutual impedance parameters between node j and node i; when the node impedance matrix satisfies the symmetry condition, this expression can be further simplified to: ,in, This represents the symmetrical equivalent expression of mutual impedance. The AC electrical distance characterizes the electrical coupling strength between node i and node j in an AC network; a smaller electrical distance indicates a stronger electrical connection between the two nodes. The calculated AC electrical distances for all node pairs are then matrixed to form the AC electrical distance matrix: The AC electrical distance matrix is ​​used for subsequent unified electrical relationship modeling and cluster partitioning of the AC / DC hybrid structure in the distribution network.

[0040] In some embodiments, calculating the DC electrical distance between each pair of DC nodes based on the DC line parameters includes: obtaining the line resistance, rated voltage, and rated capacity of each DC line, and calculating the equivalent transmission current of each DC line based on the rated voltage and rated capacity; calculating the equivalent voltage drop on each DC line based on the equivalent transmission current and the line resistance; and calculating the DC electrical distance between each pair of DC nodes based on the equivalent voltage drop and preset DC network topology path information.

[0041] In some embodiments, the line resistance, rated voltage, and rated capacity of each DC line are taken, and the equivalent transmission current of each DC line is calculated based on the rated voltage and rated capacity. Specifically, for any DC line connecting node m and node n, the basic parameters of the DC line are read, including the DC line resistance. DC line rated voltage and the allowable transmission power or rated capacity of DC lines. Based on this, the equivalent transmission current of the DC line under rated operating conditions is calculated: ,in, Indicates DC line Equivalent transmission current under rated power conditions, This indicates the permissible transmission power or rated capacity of a DC line. This indicates the rated voltage of the DC line. When it is a bipolar DC system, the rated voltage is... Take the inter-electrode voltage value.

[0042] In some embodiments, the equivalent voltage drop on each DC line is calculated based on the equivalent transmission current and the line resistance, specifically: for any DC line According to its transmission current With line resistance Calculate the voltage drop of this DC line: ,in, Indicates DC line The equivalent voltage drop generated under rated transmission current Indicates the current transmitted through a DC line. It represents the resistance of a DC line; voltage drop reflects the voltage loss characteristics caused by the line resistance during power transmission in a DC line.

[0043] In some embodiments, the DC electrical distance between each pair of DC nodes is calculated based on the equivalent voltage drop and the preset DC network topology path information. Specifically, the voltage variation characteristics of the DC line are equivalently mapped to the voltage variation characteristics of the AC line under the same power transmission conditions, wherein the DC line voltage variation satisfies: ,in, Indicates the power transmitted through a DC line. Indicates the rated voltage of the DC line. This represents the resistance of a DC line. The voltage change in an AC line under the same active power transmission conditions is expressed as: ,in, It represents the change in AC line voltage. Indicates transmission power. Indicates the reference voltage of the AC system. Indicates the resistance of the AC line. Indicates the reactance of the AC line. This represents the typical power factor angle of an AC system. Further, the reactance-resistance ratio of an AC line is defined as follows: ,in, This represents the average reactance-resistance ratio of the AC line. Substituting the above relationship, we can obtain the form of AC line voltage change: Under equivalent modeling conditions, the DC and AC lines should satisfy the equivalent voltage change relationship under the same power transmission conditions: Therefore, the equivalent impedance expression of a DC line is derived: ,in, Indicates DC line Equivalent AC impedance amplitude, Indicates the reference voltage of the AC system. Indicates the rated voltage of the DC line. Indicates the resistance of a DC line. Indicates the reactance-resistance ratio of an AC line. This represents the power factor angle. Further, by using the equivalent impedance as the electrical distance between the nodes at both ends of the DC line, we obtain the DC electrical distance: ,in, Indicates DC line The equivalent electrical distance, Indicates the reference voltage of the AC system. Indicates the rated voltage of the DC line. This indicates the per-unit value of the DC line resistance. This represents the average reactance-resistance ratio of an AC line. The DC electrical distance represents the power factor angle; it is used to characterize the equivalent influence intensity of the DC line in the unified electrical coupling model and is used to construct the unified electrical distance matrix together with the AC electrical distance.

[0044] Through the above steps, by constructing the AC node admittance matrix and further inverting it to obtain the node impedance matrix, the topological connection relationship of the AC network is uniformly mapped to a quantifiable electrical coupling relationship. On this basis, the self-impedance and mutual impedance parameters are extracted and the AC electrical distance is calculated, so that the intensity of electrical influence between nodes can be accurately characterized. At the same time, through the step-by-step derivation of DC line power, voltage drop and AC equivalent relationship, the DC line is uniformly mapped to an electrical distance expression form consistent with the AC system, thereby realizing the comparability and consistency modeling of AC and DC networks under the same measurement system. The unified electrical distance constructed in this way can more accurately reflect the true electrical coupling strength between nodes and improve the accuracy of the AC / DC hybrid distribution network zoning results.

[0045] Step S103: Based on the converter operating data, determine the converter control constraint equations, and combine the converter control constraint equations with the preset AC node power balance equations and DC node power balance equations to form a unified AC / DC power flow equation. Perform power flow calculation on the unified AC / DC power flow equation to determine the sensitivity distance between each pair of nodes in the AC / DC hybrid distribution network. Calculate the edge weights between each pair of nodes based on the AC electrical distance, the DC electrical distance, and the sensitivity distance. Calculate the fitness value corresponding to each initial node partitioning scheme based on the edge weights between each pair of nodes. Iteratively optimize each initial node partitioning scheme using a genetic algorithm until a preset convergence condition is met. Output the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network. Each initial node partitioning scheme is determined based on the network topology of the AC / DC hybrid distribution network.

[0046] In some embodiments, converter control constraint equations are determined based on the converter operating data, and these equations are combined with preset AC node power balance equations and DC node power balance equations to form a unified AC / DC power flow equation. Specifically, this involves determining the control mode identifier for each converter based on converter operating mode data, and then using a preset set of control modes... The converter control modes are classified and matched, and the control mode set includes at least PQ control mode, DC voltage control mode, DC voltage droop control mode, and combined control mode. For the k-th converter, assuming it connects AC node i and DC node j, the corresponding control constraint equations are determined based on its control mode identifier. The converter exchanges power on the AC side. , Power exchange on the DC side And satisfy the power balance relationship: ,in, This represents the active power injected by the converter into the DC side. This indicates the active power absorbed or injected by the converter from the AC side (with the injected AC system as the positive direction). This represents the active power loss of the converter; the active power loss of the converter is represented using a quadratic model: ,in, , , These represent the loss coefficients of the k-th converter, respectively. This represents the amplitude of the AC side current of the converter. Based on this, the corresponding control constraint equations are constructed according to the converter control mode. When the converter adopts the PQ control mode, its AC side power is directly set as the reference value, i.e.: ,in, This indicates the reference value for the active power of the converter. This represents the reference value of the converter's reactive power. Therefore, its control constraint equation is expressed as: When the converter adopts DC voltage control mode, its DC side voltage satisfies: ,in, This indicates the voltage at the DC node connected to the converter. This represents the DC voltage reference value. Meanwhile, reactive power can be controlled using either a setpoint control or AC voltage control. When reactive power setpoint control is used: Therefore, the control constraints can be expressed as: When using AC voltage control: The corresponding control constraints are: ,in, This indicates the amplitude of the AC bus voltage of the converter. This represents the AC voltage reference value. When the converter adopts DC voltage droop control, its control relationship is expressed as follows: ,in, This represents the droop coefficient of the k-th converter. This represents the DC power reference value. This represents the actual active power on the DC side. Therefore, its control constraint equation is: After constructing the control constraint equations for each converter, the control constraint equations for all converters are combined sequentially to form a set of control constraint equations: ,in, This represents the set of control constraint equations for the entire system's converters. This represents the control constraint equations corresponding to the k-th converter. This represents the total number of converters. Furthermore, the converter control constraint equations are combined with the AC node power balance equations and the DC node power balance equations to form a unified AC / DC power flow equation: ,in, This represents the active power imbalance at the AC node. This represents the reactive power imbalance at the AC node. This represents the active power imbalance at DC nodes. This represents the converter control constraint equations. The unified AC / DC power flow equations are used to simultaneously describe the power balance relationships of the AC network, the power balance relationships of the DC network, and the converter control constraints, thereby achieving unified modeling and joint solution of AC / DC hybrid distribution networks.

[0047] Please refer to Figure 2 In some embodiments, the step of performing power flow calculations on the AC / DC unified power flow equation to determine the sensitivity distance between each pair of nodes in the AC / DC hybrid distribution network includes: steps S201 to S203; Step S201: The power flow calculation of the AC / DC unified power flow equation is performed using the Newton-Raphson iteration method to obtain the unified state variable converged solution of the AC / DC hybrid distribution network, and a unified Jacobian matrix is ​​constructed based on the unified state variable converged solution. In some embodiments, the unified AC / DC power flow equations established for a hybrid AC / DC distribution network are solved iteratively under given initial state variables; the unified state variables include AC node voltage magnitude, AC node phase angle, DC node voltage, and converter control variables; the unified state variables are expressed as: ,in, Represents the system state variable vector. This represents the phase angle vector of the voltage at each AC node. This represents the voltage magnitude vector at each AC node. This represents the voltage vector at each DC node. This represents the equivalent control variables within the converter. The unified AC / DC power flow equation is expressed as: ,in, This represents the active power imbalance at the AC node. This represents the reactive power imbalance at the AC node. This represents the active power imbalance at DC nodes. Let represent the converter control constraint equations. The above nonlinear equations are solved using Newton's iterative method, with linearization performed in the t-th iteration: ,in, Let represent the unified Jacobian matrix corresponding to the t-th iteration. This represents the adjustment amount of the state variable. Let represent the system residual vector at the t-th iteration. The unified Jacobian matrix is ​​defined as: Through iterative updates: until the convergence condition is met: or ,in, This represents the convergence threshold for power imbalance. This represents the convergence threshold for the state variables. When the convergence condition is met, a convergent solution for the unified state variables is obtained. The unified Jacobian matrix is ​​calculated based on the convergent solution of the state variable. .

[0048] Step S202: Perform the inversion operation on the unified Jacobian matrix to obtain the sensitivity matrix including the state variables of AC nodes, DC nodes and AC / DC interface nodes; In some embodiments, the unified Jacobian matrix is ​​structurally partitioned, and its inverse matrix is ​​represented as: The matrix elements describe the linear response relationships between the state variables. The sensitivity matrix of the AC node voltage magnitude to reactive power injection disturbances at the AC node is extracted from the inverse matrix and defined as: ,in, This represents the vector of changes in the magnitude of AC node voltage. This represents the reactive power injection disturbance vector at the AC node. This represents the voltage-reactive power sensitivity matrix. The sensitivity matrix is ​​specifically defined as follows: ,in, This represents the sub-block matrix in the inverse of the unified Jacobian matrix that corresponds to the "reactive disturbance to voltage amplitude response". This represents the voltage-reactive power sensitivity matrix, and the elements of this sensitivity matrix are... This indicates the degree to which changes in reactive power injection at node j affect the voltage amplitude at node i.

[0049] Step S203: Calculate the sensitivity distance between each node pair in the AC / DC hybrid distribution network based on the sensitivity matrix.

[0050] In some embodiments, based on the sensitivity matrix For any node pair Construct the electrical distance. First, define the fundamental logarithmic sensitivity difference between nodes: ,in, This represents the basic distance between node i and node j in terms of sensitivity. This represents the self-sensitivity of node j. This represents the sensitivity of the reactive power disturbance at node j to the voltage at node i. To comprehensively consider the differences in the impact of nodes on the entire network, the sensitivity distance between node i and node j is defined as: ,in, This represents the sensitivity distance between node i and node j. Indicates the total number of communication nodes. This represents the basic sensitivity distance between node i and node k. This represents the basic sensitivity distance between node j and node k. The sensitivity distance is used to characterize the similarity of the electrical behavior response patterns of nodes in a hybrid AC / DC distribution network. The smaller the distance, the more similar the two nodes are in the system voltage-reactive response characteristics.

[0051] Please refer to Figure 3 In some embodiments, the step of calculating the edge weights between each pair of nodes based on the AC electrical distance, the DC electrical distance, and the sensitivity distance includes: steps S301 to S303; Step S301: Perform a weighted summation on each of the AC electrical distances, DC electrical distances, and sensitivity distances to obtain a unified electrical distance value for each node pair; In some embodiments, for node pairs To establish a unified electrical distance value The calculation method is as follows: ,in, Let be the electrical distance between node i and node j calculated in the AC network. The electrical distance between node i and node j, calculated equivalently via a DC line, is given. The electrical distance between nodes is calculated based on the sensitivity matrix. , , Let the weighting coefficients satisfy: ,in, The AC electrical distance is calculated from the inverse matrix Z of the AC node admittance matrix. The equivalent electrical distance for a DC line. Based on sensitivity matrix The constructed distance between nodes; where the sensitivity distance is... Defined as: , , ,in, This represents the sensitivity of the reactive power disturbance at node k to the voltage amplitude at node i. Let N represent the self-sensitivity of node k, and N be the total number of AC nodes. Through the above calculations, the unified electrical distance matrix corresponding to each node pair is obtained. .

[0052] Step S302: The unified electrical distance values ​​are superimposed with a preset non-zero regularization constant to obtain the distance correction parameters for each node pair. In some embodiments, to avoid instability caused by the denominator being zero or too small during subsequent reciprocal calculations of the uniform electrical distance, regularization is applied to the uniform electrical distance. Specifically, for each node pair... Introduce non-zero regularization constants Construct distance correction parameters The calculation method is as follows: ,in, The uniform electrical distance obtained in step S301 For a predefined regularization constant, and satisfying This is used to avoid numerical singularity. This can be set according to network size or minimum electrical distance, for example: ,in This is the adjustment coefficient. Through the above processing, a stable distance correction parameter matrix is ​​obtained. .

[0053] Step S303: Perform a reciprocal operation on the distance correction parameter to obtain the edge weights between each pair of nodes used to characterize the electrical coupling strength between nodes.

[0054] In some embodiments, for any node pair Its edge weight Defined as: ,in, For distance correction parameters, This represents the electrical coupling strength between node i and node j. To further enhance numerical stability, when When the value is extremely small, a lower limit protection mechanism can be introduced: ,in, To prevent stable constants with a denominator of zero, the edge weight matrix is ​​obtained as follows: Among them, the edge weights The larger the value, the stronger the electrical coupling between node i and node j; edge weight The smaller the value, the weaker the electrical connection between nodes. This edge weight matrix... This is used for weighted modeling and optimization calculation of the network structure during subsequent cluster partitioning.

[0055] Please refer to Figure 4 In some embodiments, the step of iteratively optimizing each of the initial node partitioning schemes using a genetic algorithm until a preset convergence condition is met, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network, includes steps S401 to S405. Step S401: Parse the adjacency matrix encoding corresponding to each initial node partitioning scheme to determine the cluster to which each node belongs in the AC / DC hybrid distribution network, and calculate the weighted modularity index corresponding to each initial node partitioning scheme based on the edge weight between each cluster and each node pair. In some embodiments, the initial node partitioning scheme obtained by chromosome encoding is represented as an adjacency matrix. ,in, Based on the adjacency matrix A, cluster analysis is performed on all nodes in the network to obtain the node cluster mapping relationship: ,in Indicates the cluster number to which node i belongs. Read the weighted adjacency matrix calculated using the unified electrical distance and sensitivity: ,in This represents the electrical coupling strength between node i and node j. Based on the above relationship, the total edge weight of the network is calculated: ; Calculate the weighted degree of each node: Define indicator functions: The final weighted modularity index is obtained as follows: ,in, The edge weight between node i and node j is used to characterize the electrical coupling strength. Let be the weighted degree of node i, representing the sum of the weights of all edges connected to that node; It is half of the total edge weight of the network, used for normalization processing; The cluster number to which node i belongs; For indicator functions, when The value is 1 if the condition is met, and 0 otherwise. This represents the total number of communication nodes. This modularity metric measures the degree of aggregation of the current partitioning scheme in terms of electrical coupling structure. A higher modularity indicates tighter internal connections and sparser cross-cluster connections.

[0056] Step S402: Based on the power operation data of each cluster and each node, calculate the source-load balance index corresponding to each initial node partitioning scheme. In some embodiments, after the cluster partitioning is completed, the power balancing capability within each cluster is further evaluated by combining the power operation data of each node. For each cluster At each scheduling time t, the difference between the active power output of all distributed power sources in the cluster and the active power demand of the load is calculated to obtain the net power of the cluster: ,in, Let be the net power of the k-th cluster at time t. The active power output of the distributed power source at node i. Let i be the active power demand of the load. Let k be the set of nodes contained in the k-th cluster. This represents the number of time sampling periods. Based on this, a source-load balance index is defined by statistically analyzing the relationship between the fluctuation level of net power at each moment and its maximum value. : ; in, The total number of clusters, The length of the time series. For cluster net power, The maximum value is taken over the time dimension. This metric is used to characterize the self-balancing capability within each cluster; a higher value indicates a higher degree of source-load matching.

[0057] Step S403: Based on the external power exchange data of each cluster and each node, calculate the interface fluctuation index corresponding to each initial node partitioning scheme, wherein the power operation data and the external power exchange data are calculated based on the converter operation data. In some embodiments, to further measure the impact of the zoning scheme on the external power grid, power exchange fluctuations between each cluster and the upper-level power grid are modeled and analyzed. For each cluster Its external exchange power at time t is defined as And calculate its average switching power over the time period: Further calculate the normalized fluctuation of the cluster interface power: ; in, This represents the external switching power of cluster k at time t. This indicates the average switching power of the cluster. Indicates the maximum allowed switching capacity of the cluster. For the number of clusters, This refers to the duration of the time interval. This metric characterizes the stability of the cluster's external power output; a smaller value indicates more stable fluctuations.

[0058] Step S404: The weighted modularity index, the source load balance index, and the interface fluctuation index are weighted and summed according to preset weight coefficients to obtain the fitness value corresponding to each initial node partitioning scheme. In some embodiments, to uniformly evaluate the merits of different initial partitioning schemes, the weighted modularity index obtained in step S401, the source-load balance index obtained in step S402, and the interface fluctuation index obtained in step S403 are uniformly integrated to construct a comprehensive fitness function. : ,in, As a weighted modularity metric, As an index of source-load balance, For interface power fluctuation indicators, This represents the normalized upper limit of the volatility indicator. , , These are weighting coefficients used to reflect the importance of different indicators in partition optimization, and they satisfy... This integrated fitness function can simultaneously consider network structure compactness, source-load self-balancing capability, and external power stability, thereby achieving a unified optimization evaluation of multiple objectives.

[0059] Step S405: Use a genetic algorithm to iteratively optimize the node partitioning schemes corresponding to each fitness value until the preset convergence condition is met, and output the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network.

[0060] In some embodiments, after obtaining the fitness values ​​corresponding to each initial node partitioning scheme, a genetic algorithm is used to globally optimize the node partitioning scheme. The probability of an individual being selected is calculated based on its fitness value. ,in, Let be the selection probability of the i-th individual. Let i be the fitness value corresponding to the i-th individual. The population size is defined by [insert population size here]. Crossover and mutation operations are performed sequentially, with crossover and mutation probabilities adaptively adjusted based on the current iteration progress to enhance global search capability and local convergence capability. After each generation of population update, fitness is recalculated, and the individual with the highest fitness is retained for the next generation to avoid losing the optimal solution. Iteration stops when any of the following termination conditions is met: the number of iterations reaches a preset maximum value. Or the fitness change over G generations is less than the threshold. Finally, the individual with the highest fitness value is selected from the population, and its corresponding adjacency matrix is... This is the optimal node partitioning scheme, and further analysis yields the cluster structure and its boundary relationships.

[0061] In some embodiments, the step of iteratively optimizing the node partitioning schemes corresponding to each fitness value using a genetic algorithm until a preset convergence condition is met, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network, includes: selecting each node partitioning scheme to obtain a set of parent individuals; performing crossover processing on the gene encoding of the parent individuals set using a simulated binary crossover operator to obtain a first set of offspring individuals; perturbing the gene loci of the first offspring individuals set using a polynomial mutation operator to obtain a second set of offspring individuals; and performing connectivity constraint verification on the second set of offspring individuals until the verification result meets the preset convergence condition, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network.

[0062] In some embodiments, a parent set of individuals is obtained by selecting each node partitioning scheme. Specifically, after calculating the fitness of each initial node partitioning scheme, a roulette wheel selection strategy is adopted to select the parent set from the current population in order to prioritize the inheritance of high-quality solutions. The fitness value of each individual is used as the basis for its selection probability. A probability distribution is constructed through fitness normalization, so that individuals with higher fitness have a greater probability of being selected. The selection probability of a parent individual is defined as: ,in, Let be the probability that the i-th individual is selected as the parent. Let be the fitness value of the i-th individual. This represents the current population size. Random sampling is performed using cumulative probability intervals to generate a set of parent individuals. Simultaneously, several individuals with the highest current fitness are retained as elite individuals to avoid losing the optimal solution, thus obtaining the set of parent individuals.

[0063] In some embodiments, a simulated binary crossover operator is used to crossover the gene codes of the parent set of individuals to obtain a first set of offspring individuals. Specifically, to enhance the continuity of the search space and the global optimization capability, a simulated binary crossover operator is used to recombine genes in the parent set of individuals, thereby generating the first set of offspring individuals. The gene code of each individual is represented in the form of an adjacency matrix, i.e.: ,in, , indicating whether node i and node j belong to the same cluster; The total number of nodes. For any pair of parent individuals and The l-th gene locus is subjected to SBX crossover to generate offspring, defined as follows: First, a random number is generated. Calculate the distribution factor : This then generates two offspring gene values: , ,in, , This represents the l-th gene locus in the parent individual; This is the cross-distribution index, used to control the range of cross-perturbations; Let be a uniform random variable. By performing the above operation dimension-wise on all gene loci, the set of the first generation individuals is obtained.

[0064] In some embodiments, a polynomial mutation operator is used to perturb the gene loci of the first offspring set to obtain a second offspring set. Specifically, to avoid the population getting trapped in a local optimum, a polynomial mutation operator is used to perturb the first offspring set, thereby enhancing population diversity and obtaining a second offspring set. For any offspring individual's l-th gene locus... First, generate random numbers. And calculate the variable-asynchronous length factor. : Then update the gene values: ,in, This represents the gene value before the mutation. This represents the gene value after mutation. It is the distribution index of variation. , These represent the upper and lower bounds of the allowed value range for a gene. Since genes are represented as 0-1 adjacency relationships in this invention, rounding mapping is required after mutation. Thus, the second set of offspring individuals that satisfy the discrete topological constraints is obtained.

[0065] In some embodiments, connectivity constraint verification is performed on the second generation set of individuals until the verification result meets the preset convergence condition. Then, the node partitioning scheme with the highest fitness value is output as the partitioning result of the AC / DC hybrid distribution network. Specifically, since the crossover and mutation processes may cause the cluster partitioning result to fail to meet connectivity constraints, constraint repair processing is required for the second generation set of individuals. Specifically, for each individual's adjacency matrix A, the following steps are performed: A subgraph structure for each cluster is constructed based on matrix A, and its connectivity is detected; if an isolated node or a non-connected subgraph exists within a cluster, the node is reassigned to the nearest neighboring cluster with the smallest electrical distance. This process can be represented as: ,in, Update the cluster number for node i. For the set of all candidate clusters, Let be the average electrical distance between node i and cluster k. After all individuals have had their constraints fixed, recalculate their fitness values ​​F and update the population structure. Determine if the algorithm meets the convergence condition: or ,in, This represents the optimal fitness of the population in generation t. The convergence threshold, This represents the maximum number of iterations. The iteration process terminates when the convergence condition is met. After the genetic algorithm converges, the individual with the highest fitness value is selected from the final population as the optimal solution, and its corresponding adjacency matrix is ​​represented as: The node partitioning relationship represented by this matrix is ​​the optimal partitioning result for the AC / DC hybrid distribution network. Further based on... The following can be obtained through parsing: the cluster number to which each node belongs. The set of nodes in each cluster The optimal partitioning scheme considers both the topological boundary structure between clusters and the cross-cluster connections. It simultaneously takes into account electrical coupling strength, source-load balancing capability, and interface power stability, thereby achieving globally optimized partitioning of the AC / DC hybrid distribution network.

[0066] Through the above steps, a unified electrical coupling relationship based on AC electrical distance, DC electrical distance and sensitivity distance is introduced into the edge weight model. A multi-objective fitness function is formed by combining modularity, source-load balance and interface power fluctuation. At the same time, a genetic algorithm is used for global iterative optimization and constraint repair, so that the partitioning results can simultaneously take into account the network topology compactness, electrical response consistency and operational stability, thereby effectively improving the accuracy and robustness of the partitioning results of AC / DC hybrid distribution networks.

[0067] It is understood that the above-described device embodiments correspond to the method embodiments of the present invention, and can implement the node partitioning method for AC / DC hybrid distribution networks provided by any of the above-described method embodiments of the present invention. More detailed workflows and principles of this system can be found, but are not limited to, in the relevant descriptions of the above methods.

[0068] 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.

[0069] Based on the above embodiments of the node partitioning method for AC / DC hybrid distribution networks, 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 node partitioning method for AC / DC hybrid distribution networks according to any embodiment of the present invention.

[0070] 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.

[0071] 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.

[0072] 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.

[0073] 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 node partitioning method of the AC / DC hybrid distribution network described in any of the above-described method embodiments of the present invention.

[0074] 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.

[0075] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A node partitioning method for an AC / DC hybrid distribution network, characterized in that, include: Acquire AC line parameters, DC line parameters, and converter operating data of the AC / DC hybrid distribution network; Calculate the AC electrical distance between each AC node pair and each node pair at the AC / DC interface based on the AC line parameters; calculate the DC electrical distance between each DC node pair based on the DC line parameters. Based on the converter operating data, the converter control constraint equations are determined. These equations are then combined with preset AC node power balance equations and DC node power balance equations to form a unified AC / DC power flow equation. Power flow calculations are performed on this unified AC / DC power flow equation to determine the sensitivity distance between each pair of nodes in the AC / DC hybrid distribution network. The edge weights between each pair of nodes are calculated based on the AC electrical distance, DC electrical distance, and sensitivity distance. The fitness values ​​corresponding to each initial node partitioning scheme are calculated based on the edge weights between each pair of nodes. The initial node partitioning schemes are iteratively optimized using a genetic algorithm until a preset convergence condition is met. The node partitioning scheme with the largest fitness value is then output as the partitioning result of the AC / DC hybrid distribution network. Each initial node partitioning scheme is determined based on the network topology of the AC / DC hybrid distribution network.

2. The node partitioning method for AC / DC hybrid distribution networks as described in claim 1, characterized in that, The calculation of the AC electrical distance between each AC node pair and between each node pair at the AC / DC interface based on the AC line parameters includes: Based on the AC line parameters, a node admittance matrix is ​​constructed to characterize the line connection relationship between nodes in the AC network. The node impedance matrix is ​​obtained by inverting the node admittance matrix. Extract the self-impedance parameters of each AC node and each AC / DC interface node, as well as the mutual impedance parameters between any pair of nodes, from the node impedance matrix. Based on the self-impedance parameters and mutual impedance parameters, the AC electrical distance between each AC node pair and the node pairs at the AC / DC interface is calculated.

3. The node partitioning method for AC / DC hybrid distribution networks as described in claim 1, characterized in that, The calculation of the DC electrical distance between each pair of DC nodes based on the DC line parameters includes: Obtain the line resistance, rated voltage, and rated capacity of each DC line, and calculate the equivalent transmission current of each DC line based on the rated voltage and rated capacity. Calculate the equivalent voltage drop on each DC line based on the equivalent transmission current and the line resistance described above. Based on the equivalent voltage drop and the preset DC network topology path information, the DC electrical distance between each pair of DC nodes is calculated.

4. The node partitioning method for AC / DC hybrid distribution networks as described in claim 1, characterized in that, The process of performing power flow calculations on the unified AC / DC power flow equations to determine the sensitivity distance between node pairs in the AC / DC hybrid distribution network includes: The power flow calculation of the AC / DC unified power flow equation is performed using the Newton-Raphson iteration method to obtain the unified state variable convergent solution of the AC / DC hybrid distribution network, and a unified Jacobian matrix is ​​constructed based on the unified state variable convergent solution. The unified Jacobian matrix is ​​inverted to obtain a sensitivity matrix including the state variables of AC nodes, DC nodes, and AC / DC interface nodes. The sensitivity distance between each node pair in the AC / DC hybrid distribution network is calculated based on the sensitivity matrix.

5. The node partitioning method for AC / DC hybrid distribution networks as described in claim 1, characterized in that, The calculation of edge weights between node pairs based on the AC electrical distance, the DC electrical distance, and the sensitivity distance includes: The AC electrical distance, DC electrical distance, and sensitivity distance are weighted and summed to obtain a unified electrical distance value for each node pair. The uniform electrical distance values ​​are superimposed with a preset non-zero regularization constant to obtain the distance correction parameters for each node pair. The reciprocal of the distance correction parameter is used to obtain the edge weights between each pair of nodes, which characterize the electrical coupling strength between nodes.

6. The node partitioning method for AC / DC hybrid distribution networks as described in claim 1, characterized in that, The step of iteratively optimizing each initial node partitioning scheme using a genetic algorithm until a preset convergence condition is met, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network, includes: The adjacency matrix encoding corresponding to each initial node partitioning scheme is parsed to determine the cluster to which each node belongs in the AC / DC hybrid distribution network. Based on the edge weights between each cluster and each node pair, the weighted modularity index corresponding to each initial node partitioning scheme is calculated. Based on the power operation data of each cluster and each node, calculate the source-load balance index corresponding to each initial node partitioning scheme. Based on the external power exchange data of each cluster and each node, the interface fluctuation index corresponding to each initial node partitioning scheme is calculated, wherein the power operation data and the external power exchange data are calculated based on the converter operation data. The weighted modularity index, the source-load balance index, and the interface fluctuation index are weighted and summed according to preset weight coefficients to obtain the fitness value corresponding to each initial node partitioning scheme. A genetic algorithm is used to iteratively optimize the node partitioning schemes corresponding to each fitness value until the preset convergence condition is met. The node partitioning scheme with the largest fitness value is then output as the partitioning result of the AC / DC hybrid distribution network.

7. The node partitioning method for AC / DC hybrid distribution networks as described in claim 6, characterized in that, The step of iteratively optimizing the node partitioning schemes corresponding to each fitness value using a genetic algorithm until a preset convergence condition is met, and then outputting the node partitioning scheme with the largest fitness value as the partitioning result of the AC / DC hybrid distribution network, includes: By selecting the partitioning scheme for each node, a set of parent individuals is obtained; The gene codes of the parent generation set are cross-processed using a simulated binary crossover operator to obtain the first generation set of offspring individuals; The gene loci of the first offspring set are perturbed using a polynomial mutation operator to obtain the second offspring set. The second set of offspring individuals is subjected to connectivity constraint verification until the verification result meets the preset convergence condition. Then, the node partitioning scheme with the largest fitness value is output as the partitioning result of the AC / DC hybrid distribution network.

8. A node zoning system for an AC / DC hybrid distribution network, characterized in that, The node partitioning system is used to execute the node partitioning method for AC / DC hybrid distribution networks as described in any one of claims 1-7.

9. A terminal device, characterized in that, It includes a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein when the processor executes the computer program, it implements the node partitioning method for a hybrid AC / DC distribution network as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that, include: A stored computer program, wherein, when the computer program is executed, it controls the device containing the computer-readable storage medium to perform the node partitioning method for an AC / DC hybrid distribution network as described in any one of claims 1-7.

Images