汽轮机通流部分故障诊断方法研究  

Study on Fault Diagnosis of Steam Turbine Flow Path Components

作　　者：董晓峰[1] 顾煜炯[1] 杨昆[1] 朝格图胡日都 

机构地区：[1]电站设备状态监测与控制教育部重点实验室(华北电力大学),北京市昌平区102206

出　　处：《中国电机工程学报》2010年第35期71-77,共7页Proceedings of the CSEE

基　　金：新世纪优秀人才支持计划项目(NCET-08-0769)~~

摘　　要：提出一种变工况下汽轮机通流部分故障诊断方法。汽轮机通流部分故障在任何一个工况下都有可能发生,这就需要故障诊断系统具有智能化的故障诊断功能,在任何工况下都能够自动提取故障征兆进行故障诊断。该文应用Q-Q图证明了稳态工况下汽轮机通流部分故障监测参数是服从正态分布的,应用正态分布的"3σ规则"来求取故障监测参数的正常工作范围,结合运行规程的规定确定故障监测参数各段工作范围的隶属度,选择合适的函数作为故障征兆的隶属度函数。计算多个工况下汽轮机通流部分故障监测参数的正常工作范围,运用曲线拟合得到工况与故障监测参数正常工作范围的映射,解决了变工况下故障征兆隶属度函数的自动获取问题。利用提取出来的通流部分故障主元特征对通流部分几种常见故障模式进行分类,形成故障模式类,对于同一故障类中的故障模式,利用粗糙集理论建立故障诊断决策表,提取对故障识别有贡献的故障征兆构建故障诊断规则,应用知识依赖度为故障诊断规则的前提条件分配权重,克服了主观分配权重存在的不足,减少了故障诊断推理过程中的不确定性影响,使诊断结果更加准确。A fault diagnosis method to identify fault of steam turbine flow path components in varialbe condition was proposed. The steam turbine flow path components fault could be happened in varialbe condition, so the fault diagnosis system must have the ability to automatically extract fault symptoms to identify the fault at any unit load. The Q-Q plot is used to verify that thermal parameters are all distributed normally under steady condition. The normal operation range of thermal parameters is obtained by using 3s rule, which combined with operation rules to obtain all kinds of operation range membership degree of thermal parameters. After then, membership function of fault symptoms are got by selecting suitable function. A mapping between normal operation range of a fault systom and working condition under multi-load is built, which to solve the problem of the fault symptom membership function’s automatic acquisition. Extracted principal component feature of fault for steam turbine flow path components is used to class several common fault modes into five fault classifications. The fault diagnosis decision table is built by using rough set theory, which is used to extract fault symptoms that are useful for identifying the fault from several fault modes included in a fault classification, then fault diagnosis rules are got. A method of weight distribution using knowledge dependency is used to distribute the power of the precondition of fault diagnosis rules, which overcomes the disadvantage of weight distribution using expert experience to decrease the uncertainty in the inference process of fault diagnosis, so the diagnosis result is more accurate.

关 键 词：汽轮机 通流部分 主元分析 聚类分析 粗糙集 加权模糊逻辑 故障诊断 

分 类 号：TK267[动力工程及工程热物理—动力机械及工程]