电力信息与通信技术2025，Vol.23Issue(4)：18-24,7.DOI:10.16543/j.2095-641x.electric.power.ict.2025.04.03
基于图计算的电力变压器状态评估及溯源研究
Research on State Evaluation and Traceability of Power Transformers Based on Graph Computing
摘要
Abstract
Accurately evaluating the operating status of power transformers is of great significance for ensuring the safe and stable operation.This paper proposes to conduct research on the state evaluation and fault tracing of power transformers based on graph calculation throughout entire life cycle.Firstly,a power transformer evaluation index system was established based on supplier production information and equipment operation;Secondly,differential scoring of evaluation indicators was achieved using quantile theory,and an objective quantitative evaluation strategy for power transformers based on graph models was proposed.Finally,on the basis of state evaluation,a fault diagnosis model based on graph convolutional neural networks was established,providing guidance and suggestions for tracing the source of transformer state degradation.The results show that the work done in this paper objectively quantifies the analysis through a multi-dimensional data linkage graph model,effectively reducing human subjective discretion in the evaluation process.At the same time,compared with deep neural networks,it improves the accuracy of fault identification under few samples,providing support for efficient operation and maintenance,fault handling of power transformers.关键词
电力变压器/图计算/客观量化评估/故障溯源/图卷积神经网络Key words
power transformer/graph computing/objective quantitative evaluation/fault tracing/graph convolutional network分类
电子信息工程引用本文复制引用
郭松,党建,王勇,吕挺,赵宁波,李昳溦,韦江瑶,霍浩伟..基于图计算的电力变压器状态评估及溯源研究[J].电力信息与通信技术,2025,23(4):18-24,7.基金项目
国家自然科学基金项目(52009106) （52009106）
国网陕西省电力有限公司科技项目(5226WZ230001). （5226WZ230001）
