基于多元两重联系数的汽轮机故障诊断  

Fault diagnosis of turbine based on multiple dual connection number

作　　者：赵玉铃[1] 赵克勤[2] 

机构地区：[1]浙江水利水电学院,杭州310018 [2]浙江大学非传统安全与和平发展中心集对分析研究所,杭州310058

出　　处：《现代制造工程》2016年第2期155-160,共6页Modern Manufacturing Engineering

基　　金：浙江省科技厅项目(2009C32109)

摘　　要：针对汽轮机故障诊断受多种不确定因素干扰问题,把集对分析中的多元两重联系数用于汽轮机故障诊断:先建立基于多元两重联系数的待诊断故障与故障知识库中各类故障的欧氏距离表达式,再对表达式中的不确定系数赋值,按所得欧氏距离的值由小到大依次排出首要故障、次故障及再次故障等,然后分析数据不确定性对故障排序的影响,得到故障诊断结论。经实例应用并与其他方法所得结论对照分析表明,基于多元两重联系数进行的欧氏距离分析计算,所得故障排序客观合理,所述方法物理意义明确,计算简便实用,不仅可用于汽轮机故障诊断,也可供其他设备或系统故障诊断时选用。Formation and diagnosis of turbine vibration fault usually is affected by many uncertain factors. Based on the uncertainty of power equipment fault diagnosis process, the uncertainty set-pair analysis connection mathematical theory in the set-pair analysis is used for fault diagnosis of turbine. First, calculate the distance between all kinds of fault diagnosis samples and the equipment fault knowledge base. List the orders of primary fault, secondary fault, and tertiary fault from small to big according to the Euclidean distances, and then analyze the influence of the uncertainty of the original data on the fault diagnosis. The conclusion of fault diagnosis is obtained. The application example shows, under the uncertainty of power equipment fault diagnosis process, by using the system uncertainty set-pair analysis connection number theory for fault diagnosis of turbine, the order of faults obtained is reasonable and practical. The physical meaning of the method presented is clear. The method is simple and practical, which can be used for fault diagnosis of other power equipment or systems, as well as the diagnosis of turbine vibration fault.

关 键 词：汽轮机 故障诊断 不确定性 集对分析 多元两重联系数 

分 类 号：TM[电气工程]