广东电力2025，Vol.38Issue(9)：44-51,8.DOI:10.3969/j.issn.1007-290X.2025.09.005
多源数据驱动的核心城区配电网风险画像与韧性提升策略
Risk Profile and Resilience Enhancement Strategies for Distribution Networks in Core Urban Areas Driven by Multi-source Data
摘要
Abstract
In those old towns with high load density,the distribution networks operate under complex conditions and face challenges in update and iteration,making accurate risk assessment crucial for urban energy security.However,traditional risk assessment methods suffer from two major limitations.One is that the systems relying on expert experience exhibit strong subjectivity,making it difficult to scientifically determine indicator weights.The other is though the data-driven methods can objectively classify the distribution transformer areas(DTAs)based on multi-dimensional historical data,but the results often lack clear risk-oriented business insights and are insufficiently interpretable.To address these issues,this paper proposes a novel method integrating data-driven and knowledge-driven approaches for risk assessment and validation in DTAs.This method constructs DTA profiles through unsupervised clustering and quantifies risk scores using a dynamic weighting model,which not only assigns quantifiable business meanings to unsupervised clustering profiles but also utilizes the intrinsic structural features of the profiles to reversely validate the objectivity and accuracy of the risk assessment model,forming a cross-verification system.Finally,the effectiveness of the proposed method is demonstrated through case studies using actual data from over 2 000 DTAs in high-load-density old towns.This study provides a reference for refined management,risk prevention,and resilience enhancement in the distribution networks.关键词
存量配电网/数据驱动/知识驱动/风险评估/画像/韧性Key words
stock distribution network/data-driven/knowledge-driven/risk assessment/profile/resilience分类
信息技术与安全科学引用本文复制引用
徐强,邱显欣,何芊慧,徐力,刘盾盾,张洪财..多源数据驱动的核心城区配电网风险画像与韧性提升策略[J].广东电力,2025,38(9):44-51,8.基金项目
广东电网有限责任公司科技项目(030121KC23120006) （030121KC23120006）
广东省基础与应用基础研究基金区域联合基金项目(2022A1515110738) （2022A1515110738）
