基于贝叶斯网络柔性多状态开关可靠性建模与分析  

Modeling and Analysis of the Reliability of Flexible Multiple State Switch Based on Bayesian Network Method

作　　者：全少理 薛俞 李秋燕 刘文霞[2] 郭勇 徐雅惠[2] QUAN Shaoli;XUE Yu;LI Qiuyan;LIU Wenxia;GUO Yong;XU Yahui(Economic and Technological Research Institute of State Grid Henan Electric Power Company,Zhengzhou 450052,China;State Key Laboratory of Alternate Electrical Power System with Renewable Energy Sources,North China Electric Power University,Beijing 102206,China)

机构地区：[1]国网河南省电力公司经济技术研究院,河南郑州450052 [2]华北电力大学新能源电力系统国家重点实验室,北京102206

出　　处：《华北电力大学学报（自然科学版）》2019年第5期25-35,共11页Journal of North China Electric Power University：Natural Science Edition

基　　金：国家重点研发计划项目(2017YFB0903100);国家电网公司科技项目(521104170043)

摘　　要：为了精确建立柔性多状态开关可靠性模型并挖掘元件可靠性的薄弱环节,提出了基于贝叶斯网络对柔性多状态开关可靠性建模的方法。首先基于不同故障特征,采用马尔科夫过程建立了多状态可靠性模型。而后,分析了MMC拓扑结构,依据故障影响关系,建立了MMC的故障树,利用故障树中的事件和门规则构造贝叶斯网络。通过算例验证模型正确性,采用大量数据仿真结合贝叶斯反向推理功能,找出了系统的薄弱环节,分析了子模块冗余度对多状态开关的影响。研究成果可为后续多状态开关在配电网中的应用提供支撑。This paper proposed a reliability modeling of flexible multi-state switch based on Bayesian network in order to excavate the weaknesses of component reliability,Firstly,this paper built a multi-state reliability model by using Markov process based on fault characteristics.Then,this paper analyzed the MMC topology,forming a fault tree of MMC according to the fault influence relationship and using its events and gate rules to construct Bayesian network.Next,this paper verified the correctness of the model by examples.The modeling results show that the model identifies the weaknesses of the system and analyzes the influence of sub-module redundancy on multi-state switch with a large number of simulation and Bayesian backward reasoning function,therefore providing support for the application of multi-state switch in distribution network.

关 键 词：柔性多状态开关 MMC 可靠性评估 贝叶斯网络 

分 类 号：TM46[电气工程—电器]