浙江电力2025，Vol.44Issue(7)：33-43,11.DOI:10.19585/j.zjdl.202507004
基于关键影响因素量化分析和HGNN的暂态稳定评估
A transient stability assessment method using quantitative analysis of key influencing factors and HGNN
摘要
Abstract
In current research on transient stability assessment,the absence of quantitative analysis of key influenc-ing factors hampers the accuracy of assessment results.Thus,a transient stability assessment method for power sys-tems using hierarchical graph neural network(HGNN)is proposed,concentrating on the quantitative analysis of two crucial factors:fault area and the proportion of renewable energy.Firstly,graph theory is employed to simplify and partition the fault areas,identifying key fault areas that significantly impact transient stability assessment.Sec-ondly,with doubly-fed asynchronous wind turbine generators as typical renew energy generation devices and consid-ering their equivalent characteristics,the influence of different renewable energy proportions on the input features of transient stability assessment is quantitatively analyzed.Then,the impact of key fault areas and renewable energy proportion on transient stability assessment is explored using the HGNN model.Finally,a case study is carried out on the IEEE 39-bus system.The results indicate that the performance indicators of the HGNN model outperform those of traditional models.Identifying key fault areas can enhance assessment accuracy,while an increase in the proportion of renewable energy will reduce assessment accuracy to a certain degree.关键词
暂态稳定评估/HGNN/关键区域识别/特征量化Key words
transient stability assessment/HGNN/critical region identification/feature quantification引用本文复制引用
杨凯璇,卢国强,傅国斌,张文朝,刘利军,李杏..基于关键影响因素量化分析和HGNN的暂态稳定评估[J].浙江电力,2025,44(7):33-43,11.基金项目
青海省重点研发与转换计划项目(2023-GX-158) （2023-GX-158）
