In this article, we suggest an extension of our proposed method in fault detection called Reduced Kernel Principal Component Analysis (RKPCA) (Taouali et al., 2015) to fault isolation. To this end, a set of structured residues is generated by using a partial RKPCA model. Furthermore, each partial RKPCA model was performed on a subset of variables to generate structured residues according to a properly designed incidence matrix. The relevance of the proposed algorithm is revealed on Continuous Stirred Tank Reactor.
AlcalaCFQinSJ (2010) Reconstruction-based contribution for process monitoring with Kernel Principal Component Analysis. Ind Eng Chem49(May): 7849–7857.
2.
ChapelleOVapnikVBousquetO. (2002) Choosing multiple parameters for support vector machines. Machine Learning46(1–3): 131–159.
3.
DongDMcAvoyTJ (1996) Nonlinear principal component analysis—Based on principal curves and neural networks. Computers and Chemical Engineering20(1): 65–78.
4.
GertlerJLiWHuangY. (1999) Isolation enhanced principal component analysis. AIChE Journal45(2): 323–334.
5.
HidenHGWillisMJThamMT. (1999) Non-linear principal components analysis using genetic programming. Computers & Chemical Engineering23(3): 413–425.
6.
HuangYGertlerJMcAvoyTJ (2000) Sensor and actuator fault isolation by structured partial PCA with nonlinear extensions. Journal of Process Control10(5): 459–469.
7.
JaffelITaoualiOElaissiI. (2013) Online prediction model based on the SVD-KPCA method. ISA Transactions52(1): 96–104.
8.
JaffelITaoualiOElaissiIMessaoudH (2014) A new online fault detection method based on PCA technique. IMA Journal of Mathematical Control and Information31(4): 487–499.
9.
JaffelITaoualiOHarkatMFMessaoudH (2015a) A fault detection index using principal component analysis and mahalanobis distance. IFAC-PapersOnLine48(21): 1397–1401.
10.
JaffelITaoualiOHarkatMFMessaoudH (2015b) Online process monitoring using a new PCMD index. The International Journal of Advanced Manufacturing Technology80(5): 947–957.
11.
KallasMMourotGMaquinDRagotJ (2015) Detection, isolation and fault estimation of nonlinear systems using a directional study. Journal of Physics: Conference Series659(1): 12032–12043.
12.
LaamiriIKhouajaAMessaoudH (2015) Convergence analysis of the alternating RGLS algorithm for the identification of the reduced complexity Volterra model. ISA Transactions55: 27–40. Available at: http://www.ncbi.nlm.nih.gov/pubmed/25442399
13.
LeeaJMYooCKChoiSW. (2004) Nonlinear process monitoring using kernel principal component analysis. Chemical Engineering Science59(1): 223–234.
14.
LiGAlcalaCFQinSJZhouD (2011) Generalized reconstruction-based contributions for output-relevant fault diagnosis with application to the Tennessee Eastman process. IEEE Transactions on Control Systems Technology19(5): 1114–1127.
15.
MessaoudHTaoualiOElaissiI (2014) Hybrid kernel identification method based on support vector regression and regularisation network algorithms. IET Signal Processing8(9): 981–989.
ScholkopfB (1998) Nonlinear component analysis as a kernel eigenvalues problem. Neural Computation10: 1299–1319.
18.
TaoualiOElaissiIMessaoudH (2012) Online identification of nonlinear system using reduced kernel principal component analysis. Neural Computing and Applications21(1): 161–169.
19.
TaoualiOElaissiIMessaoudH (2014) Dimensionality reduction of RKHS model parameters. ISA Transactions57: 205–210.
20.
TaoualiO. (2015) New fault detection method based on reduced kernel principal component analysis (RKPCA). International Journal of Advanced Manufacturing Technology.
21.
TaoualiOJaffelILahdhiriH. (2013) Process monitoring based on improved recursive PCA methods by adaptive extracting principal components. Transactions of the Institute of Measurement and Control35(8): 1024–1045.
