Comparison of Different Fault Detection Statistics Detectability in PCA

摘要：

　　Detectability of the sensor fault detection system is the basic criteria for selecting of different fault detection statistics.PVR and CVR in MPCA are compared with T2H.The complementary among them is analyzed qualitatively.Results of power plant data simulation show that PVR and CVR in MPCA can improve sensor fault detection,but T2H is still better for some sensors.Combination of PVR,C VR and T2H can improve sensor fault detectability of PCA-based fault detection system.

关键词： process monitoring fault detection Principal component analysis

作者: Tian Qiu Xiaojing Bai

作者单位: Key Laboratory of Measurement and Control for Industry Process North China Electric Power University Beijing, China

会议类型: 国际会议

会议名称: 2012 2nd International Conference on Computer Application and System Modeling(2012第二届计算机应用与系统建模国际会议)(ICCASM-2012)

会议地点: 沈阳

会议语种:英文

页码: 462-465

在线出版日期: 2012-07-27（万方平台首次上网日期，不代表论文的发表时间）