基于半Markov过程的变压器故障率分析  

Failure Rate Analysis of Transformers Based on Semi-Markov Process

作　　者：白翠粉[1] 高文胜[2] 程建伟 土松江.卡日 刘通 马仪 

机构地区：[1]国网能源研究院,北京102209 [2]清华大学电机工程与应用电子技术系,北京100084 [3]南方电网科学研究院,广州510080 [4]云南电网有限责任公司,昆明650217

出　　处：《高电压技术》2015年第12期3916-3921,共6页High Voltage Engineering

基　　金：国家高技术研究发展计划(863计划)(2011AA05A120)~~

摘　　要：设备的故障率曲线是制定维修策略的重要指标之一。为了合理估计变压器的故障率,提出了基于半Markov过程的分析模型。模型中考虑了变压器的热老化过程和随机故障过程。热老化过程根据绝缘纸平均聚合度的取值范围划分为4个状态,状态间的转移时间假设服从Weibull分布,Weibull分布的尺度参数由经验公式获得;随机故障过程考虑了由雷击和短路冲击引起的2个随机故障状态,老化状态和随机故障状态之间的转移时间假设服从指数分布,转移率由故障统计数据获得。对模型中Weibull分布的形状参数进行适当取值后,可获得符合实际统计结果的变压器故障率曲线,说明综合考虑热老化和随机故障过程的半Markov变压器故障模型能够较好地反映实际变压器的故障机制。The failure rate is an important indicator to develop maintenance strategies. We established a transformer fail- ure model based on semi-Markov process to analyze the failure rate. The model involves thermal aging process and random failure process. The thermal aging process is divided into four states based on average polymerization degree of insulating paper. The transition time between the states of thermal aging process is assumed to obey the Weibull distribu- tion, and the average transition time is obtained by empirical formula. The random failure process includes random failures caused by lightning strikes or short circuits shocks. The transition time between normal state and random failure state is assumed to be exponentially distributed, and the transition rate is obtained through failure statistics. With the vari- ation of the shape parameter of the Weibull distribution, a reasonable failure rate curve can be acquired by the transformer failure model presented. The conclusion can be drawn that the semi-Markov failure model for transformers considering both thermal aging and random failure processes can better reflect the actual transformer failure mechanisms.

关 键 词：变压器 故障率 半Markov过程 热老化 随机故障 WEIBULL分布 

分 类 号：TM407[电气工程—电器]