CN-121983945-A - Distribution network cluster division method, system, equipment and medium based on electrical distance module degree index

CN121983945ACN 121983945 ACN121983945 ACN 121983945ACN-121983945-A

Abstract

The invention discloses a distribution network cluster division method, a system, equipment and a medium based on an electrical distance module degree index, which relate to the technical field of power systems and comprise the steps of constructing a distribution network mathematical model based on acquired information; the method comprises the steps of calculating the running voltage value and the power injection state of each node, constructing a voltage-active power sensitivity matrix, calculating the electrical distance between any node pair to form an electrical distance matrix between the nodes, constructing an extended modularity evaluation index integrating electrical distance information based on the electrical distance matrix, obtaining a power distribution network cluster division result by adopting an iterative optimization strategy by taking the extended modularity evaluation index as an objective function, and carrying out connectivity verification and electrical rationality verification on the cluster division result. The method is beneficial to improving the matching degree of the cluster division result and the actual running state, and enhances the applicability of the method in power distribution networks with different scales and different structures.

Inventors

WEN XIANKUI
YE HUAYANG
ZENG PENG
CAI YONGXIANG
WANG YANG
YU JINSONG
QU YANGQUAN
HUANG CHANGMING
ZHANG GUO
ZHOU CHANGTAO
ZHANG JINSHENG
HE MINGJUN
REN SIYU
LIANG YUANSHENG
LI YANG
JING SHIYI
YI LONG
XIE ZHONGREN
Wen Qixiao
XU YUTAO
ZHOU KE
ZHANG JUNWEI
ZHANG FAN
ZHAO SHUAI
HU XUDONG
FAN QIANG

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260505
Application Date: 20251128

Claims (10)

1. A power distribution network cluster dividing method based on an electrical distance module degree index is characterized by comprising the steps of, Acquiring network structure information and operation parameter information of a power distribution network to be divided, and constructing a mathematical model of the power distribution network based on the acquired information; Calculating the operation voltage value and the power injection state of each node based on the mathematical model of the power distribution network, and constructing a voltage-active power sensitivity matrix according to the influence relationship of the calculated node voltage on the active injection of the node; Calculating the electrical distance between any node pair based on the voltage-active power sensitivity matrix to form an inter-node electrical distance matrix; based on the electric distance matrix, constructing an expansion modularity evaluation index fusing electric distance information; Combining the nodes by taking the expansion module degree evaluation index as an objective function and adopting an iterative optimization strategy to obtain a distribution network cluster division result; And carrying out connectivity verification and electrical rationality verification on the cluster division result.
2. The method for partitioning a power distribution network cluster based on an electrical distance modularity index of claim 1, wherein said constructing a mathematical model of the power distribution network based on the obtained information comprises, Modeling a topological structure of the power distribution network according to the acquired node set and the line connection relation, and generating a network structure model reflecting the connection relation between the nodes; The rated voltage, the load power and the installed capacity of the distributed power supply corresponding to each node are used as node attribute parameters and are introduced into the network structure model to form a node operation parameter model; And taking the resistance parameters and reactance parameters corresponding to each line as line attribute parameters, introducing the line attribute parameters into the network structure model, and constructing a mathematical model of the power distribution network, wherein the mathematical model comprises node parameters and line parameters.
3. The method for partitioning a power distribution network cluster based on an electrical distance module metric of claim 2, wherein said constructing a voltage-active power sensitivity matrix comprises, Setting an initial voltage state of each node based on a mathematical model of the power distribution network, and executing power flow calculation to obtain voltage amplitude values and power injection results of each node; based on a power flow calculation result, constructing a jacobian matrix of the power flow calculation, carrying out inversion processing, extracting a partial derivative relation corresponding to voltage-active power from the jacobian matrix, and obtaining an influence relation of each node voltage on active injection of each node; And arranging the influence relation of the voltage on active injection according to the node number to form a voltage-active power sensitivity matrix.
4. The method for partitioning a power distribution network cluster based on an electrical distance modularity index of claim 3, wherein said forming an inter-node electrical distance matrix comprises, Based on a network structure model of the power distribution network, and combining the electrical association relationship between the nodes represented in the voltage-active power sensitivity matrix, determining a corresponding electrical connection path between any two nodes; Accumulating the resistance parameters of each line on the path to obtain an equivalent resistance value between corresponding node pairs; and filling the equivalent resistance value serving as an electrical distance measurement between the nodes according to the node numbering sequence to form an electrical distance matrix between the nodes.
5. The method for partitioning a power distribution network cluster based on an electrical distance module metric of claim 4, wherein said constructing an extended module metric for fusing electrical distance information comprises, Based on the electric distance matrix, calculating a corresponding edge weight for any node pair, and introducing the edge weight into a network connection relation; Setting a resolution parameter for adjusting the cluster scale, and introducing the resolution parameter into a modularity calculation process; And constructing an extended modularity evaluation index fusing the electrical distance information based on the side weight, the resolution parameter and the node weighting information.
6. The method for power distribution network cluster division based on the electrical distance module degree index of claim 5, wherein the step of obtaining the power distribution network cluster division result comprises the steps of, Each node in the power distribution network is respectively set as an independent cluster; Based on the expansion module degree evaluation index, cluster merging operation is carried out on the adjacent nodes, and the cluster structure is updated after each round of processing; And when the convergence condition is met, determining the current cluster structure as a final distribution network cluster division result.
7. The method for partitioning a power distribution network cluster based on an electrical distance modularity index of claim 6, wherein said performing connectivity verification and electrical rationality verification on said clustered partitioning result comprises, Traversing node connection relations in each cluster based on a network structure model of the power distribution network, and judging whether each cluster forms a connected sub-network or not; based on the electric distance matrix, respectively calculating average electric distances among nodes in the same cluster; And calculating the average electrical distance between the nodes of different clusters based on the electrical distance matrix, and comparing the average electrical distance with the average electrical distance in the corresponding clusters.
8. The distribution network cluster division system based on the electrical distance module degree index is applied to the distribution network cluster division method based on the electrical distance module degree index according to any one of claims 1-7, and is characterized by comprising a model construction module, a matrix construction module, an evaluation index construction module, a cluster division module and a verification module; the model construction module is used for acquiring network structure information and operation parameter information of the power distribution network to be divided and constructing a mathematical model of the power distribution network based on the acquired information; The matrix construction module is used for calculating the operation voltage value and the power injection state of each node based on the mathematical model of the power distribution network, constructing a voltage-active power sensitivity matrix according to the influence relation of the calculated node voltage on the active injection of the node, and calculating the electrical distance between any node pair based on the voltage-active power sensitivity matrix to form an inter-node electrical distance matrix; The evaluation index construction module is used for constructing an expansion module degree evaluation index fusing the electrical distance information based on the electrical distance matrix; the cluster dividing module takes the expansion module degree evaluation index as an objective function, and adopts an iterative optimization strategy to combine the nodes so as to obtain a power distribution network cluster dividing result; And the verification module is used for carrying out connectivity verification and electrical rationality verification on the cluster division result.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of a distribution network cluster division method based on an electrical distance modularity index as claimed in any one of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of a distribution network cluster division method based on an electrical distance modularity index as claimed in any one of claims 1 to 7.

Description

Distribution network cluster division method, system, equipment and medium based on electrical distance module degree index Technical Field The invention relates to the technical field of power systems, in particular to a power distribution network cluster division method, system, equipment and medium based on an electric distance modularity index. Background With the deep promotion of the construction of a novel power system, the permeability of distributed new energy, particularly distributed photovoltaic, in a power distribution network is continuously improved, and the traditional power distribution network is transformed into a novel power distribution network with forward bidirectional tide and multi-source access. The novel power distribution network has the characteristics of complex topological structure, various energy forms, numerous regulation and control objects and the like, and provides brand new challenges for planning, running and control of the power distribution network. Cluster division is used as an effective means for simplifying the control of a complex power distribution network, and a layered control strategy of 'zero integration' is realized by classifying nodes with similar geographic positions or electrical characteristics. The conventional power distribution network cluster division method mainly has the technical problems that although the traditional community discovery algorithm can fully consider the topological connection relation of a power distribution network, complex electric operation indexes are difficult to directly introduce, the electric coupling degree among nodes cannot be accurately reflected, the clustering analysis algorithm can process high-dimensional electric data, inherent topological structure characteristics of the power distribution network are ignored, the electrical non-connected clusters possibly appear in the division result, the conventional method lacks self-adaptive processing capacity for clusters with different densities, and the cluster scale is difficult to flexibly adjust according to the power distribution network scale and the new energy permeability. These problems result in a mismatch between the cluster division results and the actual operating characteristics of the distribution network, affecting the effectiveness of subsequent optimization control. Disclosure of Invention In view of the above problems, the invention provides a power distribution network cluster division method, system, equipment and medium based on an electric distance modularity index. Therefore, the invention aims to solve the problems that the existing method lacks self-adaptive processing capability for clusters with different densities, and is difficult to flexibly adjust the cluster scale according to the power distribution network scale and the new energy permeability. The power distribution network cluster dividing method based on the electrical distance module degree index comprises the steps of obtaining network structure information and operation parameter information of a power distribution network to be divided, constructing a power distribution network mathematical model based on the obtained information, calculating operation voltage values and power injection states of all nodes based on the power distribution network mathematical model, constructing a voltage-active power sensitivity matrix according to the influence relation of calculated node voltages on node active injection, calculating electrical distances between any node pairs based on the voltage-active power sensitivity matrix to form an electrical distance matrix between the nodes, constructing an expansion module degree evaluation index integrating the electrical distance information based on the electrical distance matrix, combining the nodes by adopting an iterative optimization strategy based on the expansion module degree evaluation index as a target function to obtain a power distribution network cluster dividing result, and conducting connectivity verification and electrical rationality verification on the cluster dividing result. The method for dividing the power distribution network cluster based on the electrical distance module degree index comprises the steps of constructing a power distribution network mathematical model based on acquired information, modeling a topology structure of the power distribution network according to acquired node sets and line connection relations, generating a network structure model reflecting the connection relations among nodes, introducing rated voltage, load power and distributed power supply installed capacity corresponding to each node into the network structure model as node attribute parameters to form a node operation parameter model, and introducing resistance parameters and reactance parameters corresponding to each line into the network structure model to construct the power distribution network mathematical model containing the node pa