基于映射向量和加权Q贡献图的故障识别研究  

Research on Fault Identification by Mapping Vector and Weighted Q Contribution Plot

作　　者：李荣雨[1] 杨小健[1] 殷绚[1] 

机构地区：[1]南京工业大学电子与信息工程学院,南京210009

出　　处：《化工自动化及仪表》2010年第12期48-50,共3页Control and Instruments in Chemical Industry

摘　　要：在基于主元分析(PCA)的统计过程监控实施中,贡献图是一个常用的故障识别工具,但难以实现故障精确定位的缺点限制了贡献图的进一步推广。通过引入基于真实值的故障模型,推导出故障映射向量幅值与故障贡献准确性的关系,指出了传统Q贡献图的理论缺陷,并提出了一种加权Q贡献图故障识别法。在CSTR上的仿真证明,相比传统Q贡献图,该法可以更准确地识别传感器,消除了对操作员的误导。Contribution plot is a popular tool for fault identification in statistical process monitoring based on principal component analysis（PCA）,although this method can not isolate the fault sensor accurately.The theory deficiency of contribution plot was studied by introducing fault model based on true value and fault mapping vector,and an improved method of weighted Q contribution plot was proposed to enhance the fault identification capability.Simulations on continuously stirred tank reactor（CSTR） show that the weighted Q contribution method is more effective than traditional contribution plot.

关 键 词：主元分析 统计过程监控 故障识别 贡献图 

分 类 号：TP206[自动化与计算机技术—检测技术与自动化装置]