电工技术学报  2022, Vol. 37 Issue (3): 667-675    DOI: 10.19595/j.cnki.1000-6753.tces.210070
电力系统与综合能源 |
变压器故障样本多维诊断及结果可信度分析
李典阳1,2, 张育杰1,3, 冯健1, 王善渊1
1.东北大学信息科学与工程学院 沈阳 110819;
2.国网辽宁省电力有限公司 沈阳 110006;
3.国网新乡供电公司 新乡 453005
Multi-Dimensional Diagnosis of Transformer Fault Sample and Credibility Analysis
Li Dianyang1,2, Zhang Yujie1,3, Feng Jian1, Wang Shanyuan1
1. Information Science and Engineering College Northeastern University Shenyang 110819 China;
2. State Grid Liaoning Electric Power Limited Company Shenyang 110006 China;
3. State Grid Xinxiang Electric Power Supply Company Xinxiang 453005 China
全文: PDF (988 KB)   HTML
输出: BibTeX | EndNote (RIS)      
摘要 电力变压器作为电力系统的核心设备,其安全稳定运行对于电力系统具有重要意义。电力变压器在线故障诊断是实现电力变压器实时状态分析的重要方法,油中溶解气体分析是最常用的电力变压器在线故障诊断方法。目前变压器故障诊断征兆优选多采用基于启发式算法的策略,虽然相较于遍历型算法简化了筛选流程,但仍需消耗大量算力。电力变压器融合故障诊断方面的研究多注重于整体诊断效果的提升,未关注单个样本诊断结果可靠性方面的分析。为解决上述问题,该文提出一种结合数据分布的征兆自主离散及征兆自主降维优选、单事件多模型融合分析的变压器状态分析方法。经实例验证,该方法可以有效分析各征兆数据分布,进行征兆优选,可以从单个事件的角度给出变压器运行状态及可信度。
服务
把本文推荐给朋友
加入我的书架
加入引用管理器
E-mail Alert
RSS
作者相关文章
李典阳
张育杰
冯健
王善渊
关键词 优化离散征兆选择可信度辅助分析    
Abstract:Power transformer is the core equipment of power system, and its safe and stable operation is of great significance to power system. Online fault diagnosis of power transformers is an important method to realize real-time status analysis of power transformers. At present, the selection of features subset for transformer fault diagnosis mainly adopts heuristic-based method, which simplifies the selection process compared to traversal algorithms, but it still consumes a lot of computing power. Moreover, the research of hybrid fault diagnosis motheds for power transformer focuses on the improvement of the diagnosis effect in all samples, and does not pay attention to the credibility analysis of the diagnosis results in single sample. In order to solve this problem, this paper proposed a transformer state analysis method that combines autonomous discretization and optimization of data distribution signs, and single-event multi-model fusion analysis. Proved by examples, this method can effectively analyze the data distribution of each feature, perform feature optimization, and can get the operating status and reliability of the transformer from the perspective of a single event.
Key wordsOptimized discrete    feature selection    result credibility    auxiliary analysis   
收稿日期: 2021-01-15     
PACS: TM41  
基金资助:国家自然科学基金资助项目(61673093)
通讯作者: 张育杰 男,1996年生,硕士研究生,研究方向为电力设备故障诊断与状态预判。E-mail:zyj_neu@163. com   
作者简介: 李典阳 男,1987年生,博士研究生,研究方向为电网运行及电力调度控制技术与应用。E-mail:2824804703@qq. com
引用本文:   
李典阳, 张育杰, 冯健, 王善渊. 变压器故障样本多维诊断及结果可信度分析[J]. 电工技术学报, 2022, 37(3): 667-675. Li Dianyang, Zhang Yujie, Feng Jian, Wang Shanyuan. Multi-Dimensional Diagnosis of Transformer Fault Sample and Credibility Analysis. Transactions of China Electrotechnical Society, 2022, 37(3): 667-675.
链接本文:  
https://dgjsxb.ces-transaction.com/CN/10.19595/j.cnki.1000-6753.tces.210070          https://dgjsxb.ces-transaction.com/CN/Y2022/V37/I3/667