QQ客服 官方微博 反馈留言

	考虑量测错误和多源异质信息融合的变电站概率故障诊断:统一的Bayes网络模型
	牟佳男 ; 孙宏斌 ; 吴文传 ; 郭庆来 ; 张伯明 ; MU Jianan ; SUN Hongbin ; WU Wenchuan ; GUO Qinglai ; ZHANG Boming
	2016-03-30 ; 2016-03-30
关键词	Bayes网络 变电站 故障诊断 量测信道错误 多源异质信息 智能电网 superheated steam temperature active disturbance rejection control fruit fly optimization robustness TM63
其他题名	Probabilistic substation fault diagnosis considering instrumentation channels and heterogeneous information:A unified Bayesian network model
中文摘要	故障诊断对提升电网故障分析处理能力有重要意义。在变电站本地进行故障诊断可以更加充分、敏捷地利用本地的丰富信息。为适应智能电网环境下信息丰富、分布自治的趋势,提出一种基于Bayes网络的变电站故障诊断方法,考虑了保护、开关的拒误动和量测信道错误等不确定性,并综合利用顺序事件(sequence of events,SOE)、故障录波、保护闭锁等多源异质信息。使用经典的联结树(junction tree)算法进行概率推理。结果表明:与不考虑量测信道不确定性的方法相比,该方法可以更加充分利用开关SOE等信息;故障录波和闭锁等非SOE信息的引入可以有效提升诊断的准确率。; Fault diagnosis plays an important role in improving power network troubleshooting capability.Conducting local fault diagnosis in substations can make full and quick use of abundant local information.A substation fault diagnosis method was presented based on Bayesian network to adapt to a wealth of information and distributed autonomy under smart grid environment.The method is effective in dealing with uncertainties induced by malfunctions of protection relays and circuit breakers as well as instrumentation channel errors. Heterogeneous information available in substations,such as sequence of events(SOEs),wave records and blocking signals,was also integrated into the method,with the junction tree algorithm used to conduct probabilistic reasoning.The results show that breaker SOEs can be better utilized when considering instrumentation channels.Furthermore,introduction of wave records and blocking signals can significantly improve diagnosis accuracies.
语种	中文 ; 中文
内容类型	期刊论文
源URL	[http://ir.lib.tsinghua.edu.cn/ir/item.do?handle=123456789/142646]
专题	清华大学
推荐引用方式 GB/T 7714	牟佳男,孙宏斌,吴文传,等. 考虑量测错误和多源异质信息融合的变电站概率故障诊断:统一的Bayes网络模型[J],2016, 2016.
APA	牟佳男.,孙宏斌.,吴文传.,郭庆来.,张伯明.,...&ZHANG Boming.(2016).考虑量测错误和多源异质信息融合的变电站概率故障诊断:统一的Bayes网络模型..
MLA	牟佳男,et al."考虑量测错误和多源异质信息融合的变电站概率故障诊断:统一的Bayes网络模型".(2016).

个性服务

查看访问统计

相关权益政策

暂无数据

收藏/分享

除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。