TY - GEN
T1 - Partial Least Squares Regression for Recursive System Identification
AU - QIN, S. Joe
PY - 1993/12
Y1 - 1993/12
N2 - Industrial processes usually involve a large number of variables, many of which vary in a correlated manner. To identify a process model which has correlated variables, an ordinary least squares approach demonstrates ill-conditioned problem and the resulting model is sensitive to changes in sampled data. In this paper, a recursive partial least squares (PLS) regression is used for on-line system identification and circumventing the ill-conditioned problem. The partial least squares method is used to remove the correlation by projecting the original variable space to an orthogonal latent space. Applications of the proposed algorithm to a chemical process modeling problem is discussed.
AB - Industrial processes usually involve a large number of variables, many of which vary in a correlated manner. To identify a process model which has correlated variables, an ordinary least squares approach demonstrates ill-conditioned problem and the resulting model is sensitive to changes in sampled data. In this paper, a recursive partial least squares (PLS) regression is used for on-line system identification and circumventing the ill-conditioned problem. The partial least squares method is used to remove the correlation by projecting the original variable space to an orthogonal latent space. Applications of the proposed algorithm to a chemical process modeling problem is discussed.
UR - http://www.scopus.com/inward/record.url?scp=0027744929&partnerID=8YFLogxK
U2 - 10.1109/cdc.1993.325671
DO - 10.1109/cdc.1993.325671
M3 - Conference paper (refereed)
SN - 0780312988
T3 - Proceedings of the IEEE Conference on Decision and Control
SP - 2617
EP - 2621
BT - Proceedings of 32nd IEEE Conference on Decision and Control
PB - Institute of Electrical and Electronics Engineers
T2 - 32nd IEEE Conference on Decision and Control
Y2 - 15 December 1993 through 17 December 1993
ER -