文章摘要
李永刚,王罗,李俊卿,等.基于多源信息融合的同步发电机转子绕组匝间短路故障识别[J].电力系统自动化. DOI: 10.7500/AEPS20180807001.
LI Yonggang,WANG Luo,LI Junqing, et al.Identification of Inter-turn Short-circuit Fault in Synchronous Generator Rotor Windings Based on Multi-source Information Fusion[J].Automation of Electric Power Systems. DOI: 10.7500/AEPS20180807001.
基于多源信息融合的同步发电机转子绕组匝间短路故障识别
Identification of Inter-turn Short-circuit Fault in Synchronous Generator Rotor Windings Based on Multi-source Information Fusion
DOI:10.7500/AEPS20180807001
关键词: 同步发电机  匝间短路  多源信息融合  故障识别
KeyWords: synchronous generator  inter-turn short circuit  multi-source information fusion  fault identification
上网日期:2019-07-09
基金项目:中央高校基本科研业务费项目(2017XS115)
作者单位E-mail
李永刚 华北电力大学新能源电力系统国家重点实验室 lygzxm0@163.com 
王罗 华北电力大学新能源电力系统国家重点实验室 wangluo1029@163.com 
李俊卿 华北电力大学新能源电力系统国家重点实验室 junqing03@163.com 
马明晗 华北电力大学新能源电力系统国家重点实验室 NCEPU_MMH@outlook.com 
摘要:
      转子绕组匝间短路是同步发电机中一种较为常见的故障,随着故障的发展会对发电机的安全运行造成威胁。针对该故障在早期时不易检测的情况,将多源信息融合理论应用到同步发电机的转子绕组匝间短路故障识别中。根据同步发电机特点及传感器情况分析并选择合适的短路故障特征量作为证据体。将发电机中的多组故障特征证据体依据证据理论进行融合,得到高置信度的故障判定结论。同时进行同步发电机故障实验,与单特征量作对比验证了多源信息融合在发电机转子绕组匝间故障识别中的有效性。结果表明,该方法减少了单一传感器不确定性的影响,提高了故障识别准确性。
Abstract:
      Rotor winding inter-turn short circuit is a common fault in synchronous generators. With the development of the fault, it will pose a threat to the safe operation of synchronous generator. In view of the fact that the fault is difficult to detect in the early stage, multi-source information fusion is applied to the identification of inter-turn short circuit fault of synchronous generator rotor windings. According to the sensor condition and characteristics of synchronous generator, analyze and select the appropriate short-circuit fault feature quantity as the evidence body. Based on the evidence theory, the fault characteristics of generators are fused and the high confidence fault diagnosis conclusion is obtained. The synchronous generator fault experiment is carried out and the validity of multi-source information fusion in generator rotor winding inter-turn fault identification is verified by comparing with single characteristic quantity. The results show that this method reduces the influence of single sensor uncertainty and improves the accuracy of fault identification.
查看全文   查看/发表评论  下载PDF阅读器