A power system distributed adjustable resource node clustering method and system

By using global preprocessing and high-dimensional feature matrix construction, combined with FCM algorithm and fuzzy decision-making, the traditional distribution network clustering method is able to address the shortcomings in characterizing the physical connectivity and uncertain fluctuations between nodes, thus achieving refined aggregation analysis of heterogeneous resources in the distribution network.

CN122333010APending Publication Date: 2026-07-03NORTH CHINA UNIVERSITY OF TECHNOLOGY +3
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
NORTH CHINA UNIVERSITY OF TECHNOLOGY
Filing Date
2026-04-02
Publication Date
2026-07-03

AI Technical Summary

Technical Problem

Traditional distribution network clustering methods struggle to accurately characterize the physical connectivity and inherent correlations of uncertain fluctuations between nodes, leading to the invalidation of the physical and characteristic meanings of the clustering results and failing to fully tap the aggregation potential between nodes.

Method used

A high-dimensional feature matrix is ​​constructed by generating a basic feature matrix and node symbol vectors through global preprocessing. Then, hierarchical clustering with physical adjacency constraints is performed by combining the FCM algorithm and the fuzzy decision algorithm to optimize the upper limit of the number of clusters and ensure the topological connectivity and uncertainty matching of the clustering results.

Benefits of technology

It achieves the global optimal matching of fluctuation direction and fluctuation amplitude between nodes, solves the "enclave" problem caused by neglecting physical connections in traditional clustering methods, and provides a refined aggregation analysis model for heterogeneous resources in distribution networks.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122333010A_ABST
    Figure CN122333010A_ABST
Patent Text Reader

Abstract

This invention discloses a method and system for clustering distributed adjustable resource nodes in a power system. The method includes: performing global preprocessing based on the global load, topology, and node types of the power system to generate a basic feature matrix containing various load characteristics, a topological region ID defining the topological affiliation of nodes, and a node symbol vector defining the uncertainty direction of nodes; segmenting the topology of the power system based on the topological region ID and constructing a high-dimensional feature matrix within each region; running a first-stage evaluation algorithm based on the high-dimensional feature matrix to determine the upper limit of the number of clusters; and performing a second-stage clustering and evaluation based on the upper limit of the number of clusters to obtain the optimal clustering result. This invention achieves globally optimal matching of the "fluctuation direction" and "fluctuation amplitude" between nodes while ensuring clustering accuracy, providing a novel and refined modeling paradigm for the aggregation analysis of heterogeneous resources in distribution networks.
Need to check novelty before this filing date? Find Prior Art