Matchable-Observable Linear Models and Direct Filter Tuning: An Approach to Multivariable Identification

Author(s)

Romano, Rodrigo Alvite ; Pait, Felipe

• Document type: Journal article
• Source: IEEE Transactions on Automatic Control; v. 62, n. 5, p. 2180-2193, MAY 2017.
• Web of Science Citations: 8

Abstract

Identification of linear time-invariant multivariable systems can best be understood as comprising three separate problems: selection of system model structure, filter design, and parameter estimation itself. Approaching the first using matchable-observable models originally developed in the adaptive control literature and the second via direct or derivative-free optimization, effective least-squares algorithms can be used for parameter estimation. The accuracy, robustness and moderate computational demands of the methods proposed are demonstrated via simulations with randomly generated models and applied to identification using real process data. The results obtained are comparable or superior to the best results obtained using standard implementations of the algorithms described in the literature. (AU)

• FAPESP's process: 12/03719-3 - A new approach to the estimation of multivariable linear models
• Grantee: Rodrigo Alvite Romano
• Support Opportunities: Regular Research Grants