Title :
Virtual-metrology-based FDC scheme
Author :
Hsieh, Yao-Sheng ; Cheng, Fan-Tien ; Yang, Haw-Ching
Author_Institution :
Inst. of Manuf. Inf. & Syst., Nat. Cheng Kung Univ., Tainan, Taiwan
Abstract :
Most conventional fault-detection-and-classification (FDC) approaches are to find out the target devices (TDs) required for monitoring and the TDs´ related key parameters needed to be monitored, and then by applying the statistical-process-control (SPC) approach to detect the faults. However, in practical situation, an abnormal key-parameter value may not be solely caused by its own TD; instead, it may also be resulted from the other related parameters. As a result, accurate fault classification or diagnosis may not be achieved. The purpose of this paper is to propose a virtual-metrology-based baseline FDC scheme that includes the TD baseline model and FDC logic. By applying this baseline FDC scheme, fault diagnosis can then be accomplished.
Keywords :
fault diagnosis; maintenance engineering; statistical analysis; virtual instrumentation; fault detection and classification; fault diagnosis; statistical process control; target device monitoring; virtual metrology based FDC; Control charts; Data models; Indexes; Maintenance engineering; Monitoring; Predictive models; Valves; Baseline fault-detection-and-classification (FDC) scheme; automatic virtual metrology (AVM); dynamic-moving-window (DMW) scheme; keep-important-sample (KIS) scheme;
Conference_Titel :
Automation Science and Engineering (CASE), 2012 IEEE International Conference on
Conference_Location :
Seoul
Print_ISBN :
978-1-4673-0429-0
DOI :
10.1109/CoASE.2012.6386371