Two adaptive principal component analysis methods are improved based on adaptive extracting principal components (PCs) in process monitoring: recursive PCA (RPCA) and moving window PCA (MWPCA). An adaptive extracting PC algorithm is proposed using the threshold method based on the score rule in sport games to determine the number of PCs in real time. It can effectively overcome the shortcomings of the conventional cumulative percent variance method in obtaining the number of PCs. Moreover, two improved RPCA and MWPCA methods are proposed using the new threshold method to monitor an industrial process online. Similary to the forgetting factor in RPCA, an optimal variable moving window size is selected, adding forgetting factors into the data samples and covariance matrices, respectively. The results show the validity of improvements compared with the original RPCA and MWPCA in Tennessee Eastman process monitoring.
ChenJHLiuKC (2002) On-line batch process monitoring using dynamic PCA and dynamic PLS models. Chemical Engineering Science57: 63–75.
2.
ChenJX (2004) Research of Multiscale PCA with Application to Multivariate Statistical Process Monitoring. Beijing: Beijing University of Chemical Technology.
3.
ChiangLHRussellELBraatzRD (2001) Fault Detection and Diagnosis in Industrial Systems. Springer: London.
4.
DashSRengaswamyRVenkatasubramanianV (2003) Fuzzy-logic based trend classification for fault diagnosis of chemical processes. Computers and Chemical Engineering27: 347–362.
5.
DayalBSMacGregorJF (1997) Recursive exponentially weighted PLS and its applications to adaptive control and prediction. Journal of Process Control7: 169–179.
6.
DianaGTommasiC (2002) Cross-validation methods in principal component analysis: a comparison. Statistical Methods & Applications11: 71–82.
7.
DownsJJVogelEF (1993) A plant-wide industrial process control problem. Computers and Chemical Engineering17: 245–255.
8.
GertlerJCaoJ (2004) PCA-based fault diagnosis in the presence of control and dynamics. AIChE Journal50: 388–402.
9.
HeXBYangYP (2008) Variable MWPCA for adaptive process monitoring. Industrial & Engineering Chemistry Research47: 419–427.
10.
HellandKBerntsenHEBorgenOS. (1991) Recursive algorithm for partial least squares regression. Chemometrics and Intelligent Laboratory Systems14: 129–137.
JengJC (2010) Adaptive process monitoring using efficient recursive PCA and moving window PCA algorithms. Journal of the Taiwan Institute of Chemical Engineers41: 475–481.
13.
JiangZJHeXBYangYP (2008) Adaptive time-varying process monitoring basing on fast MWPCA. Journal of Nanchang University (Natural Science)32: 198–204.
14.
LiWYueHValleSQinSJ (2000) Recursive PCA for adaptive process monitoring. Journal of Process Control10: 471–486.
15.
LuoGZhangSNiuY (2005) Implementation of simulation system based on TE process. Journal of Nanjing University of Technology (Natural Science Edition)27: 77–80.
16.
LuoRMisraMHimmelblauDH (1999) Sensor fault detection via multiscale analysis and dynamic PCA. Industrial & Engineering Chemistry Research38(4): 1489–1495.
17.
MacGregorJFKourtiT (1995) Statistical process control of multivariate processes. Control Engineering Practice3: 403–414.
18.
QinSJ (1998) Recursive PLS algorithms for adaptive data modeling. Computers & Chemical Engineering22: 503–514.
19.
QinSJValleSPiovosoMJ (2001) On unifying multiblock analysis with application to decentralized process monitoring. Journal of Chemometrics15: 715–742.
RomagnoliJPalazogluA (2006) Introduction to Process Control. Boca Raton, FL: CRC Press.
22.
ÜndeyCTataraEÇınarA (2003) Real-time batch process supervision by integrated knowledge-based systems and multivariate statistical methods. Engineering Applications of Artificial Intelligence16: 555–566.
23.
VenkatasubramanianVRengaswamyRKavuriSNYinK (2003a) A review of process fault detection and diagnosis (Part I): quantitative model-based methods. Computers and Chemical Engineering27: 293–311.
24.
VenkatasubramanianVRengaswamyRKavuriSN (2003b) A review of process fault detection and diagnosis (Part II): qualitative models and search strategies. Computers and Chemical Engineering27: 313–326.
25.
VenkatasubramanianVRengaswamyRYinKKavuriSN (2003c) A review of process fault detection and diagnosis (Part III): process history based methods. Computers and Chemical Engineering27: 327–346.
26.
WangXKrugerUIrwinGW (2005) Process monitoring approach using fast moving window PCA. Industrial & Engineering Chemistry Research44: 5691–5702.
27.
WangXKrugerULennoxB (2003) Recursive partial least squares algorithms for monitoring complex industrial processes. Control Engineering Practice11: 613–632.
28.
WesterhuisJAKourtiTMacgregorJF (1998) Analysis of multiblock and hierarchical PCA and PLS models. Journal of Chemometrics12: 301–321.
29.
WiseBMGallagherNB (1996) The process chemometrics approach to process monitoring and fault detection. Journal of Process Control6: 329–348.
30.
WoldH (1966) Nonlinear estimation by iterative least squares procedures. In: DavidF (ed.) Research Papers in Statistics. New York: Wiley.
31.
WoldSGeladiPEsbensenK. (1987) Multi-way principal components-and PLS-analysis. Journal of Chemometrics1: 41–56.
32.
YangQTianFWangD (2010) Online approach of fault diagnosis based on lifting wavelets and moving window PCA. Proceedings of the 8th World Congress on Intelligent Control and Automation, Jinan, China, 2010.
33.
ZhuMGhodsiA (2006) Automatic dimensionality selection from the scree plot via the use of profile likelihood. Computational Statistics & Data Analysis51: 918–930.
