基于改进K-Means聚类算法和最小生成树融合模型的配电网规划研究

Research on Distribution Network Planning Based on Improved K-Means Clustering Algorithm and Minimum Spanning Tree Fusion Model

摘要: 本文针对配电网在规划分区时各分区用户负荷差值较大而造成可靠性不足和经济损失等问题，分别提出了一种基于负荷因子的平衡负荷初步分区方法和基于欧式距离权重的改进K-means聚类算法，并将其应用于配电网规划研究。以我国某城市街道为例，试验结果表明改进的K-means聚类在负荷分区的均匀分布上优于传统的K-means聚类方法，改进的K-means聚类负荷差值的平均值由0.364MW下降到0.1675 MW。输出聚类结果后的最短路径以实现配电网规划线路图。

Abstract: This article proposes a preliminary balanced load partitioning method based on load factors and an improved K-means clustering algorithm based on Euclidean distance weights to address the issues of insufficient reliability and economic losses caused by large differences in user loads between different zones in distribution network planning. These methods are applied to research on distribution network planning. Taking a certain urban street in China as an example, the experimental results show that the improved K-means clustering is superior to the traditional K-means clustering method in terms of uniform distribution of load zones. The average value of the improved K-means clustering load difference decreases from 0.364 MW to 0.1675 MW. The shortest path after clustering results is output to achieve distribution network planning circuit diagram.