Nonlinear predictive control of irregularly sampled data systems using identified observers. (English) Zbl 1223.93055

Findeisen, Rolf (ed.) et al., Assessment and future directions of nonlinear model predictive control. Selected papers based on the presentations at the workshop (NMPC05), Freudenstadt-Lauterbad, Germany, August 26–30, 2005. Berlin: Springer (ISBN 978-3-540-72698-2/pbk). Lecture Notes in Control and Information Sciences 358, 141-149 (2007).
From the introduction: This work aims at the identification of a nonlinear fast rate model and a nonlinear multi-rate time varying state observer from irregularly sampled (multi-rate) data, which is corrupted with unmeasured disturbances and measurement noise. The deterministic and stochastic components of the proposed model have Weiner structure. The linear dynamic component of these models is parameterized using Generalized Orthonormal Basis Filters (GOBF) [S. C. Patwardhan and S. L. Shah, From data to diagnosis and control using generalized orthonormal basis filters. Part I: Development of state observers, J. Process Control 15, No. 7, 819–835 (2005); B. Ninness and F. Gustafsson, IEEE Trans. Autom. Control 42, No. 4, 515–521 (1997; Zbl 0874.93034)]. We then proceed to show how the identified models can be used for inter-sample inferential estimation of the slowly sampled variable and also for model predictive control of such irregularly sampled, multi-rate systems. The efficacy of the proposed modeling and control scheme is demonstrated by conducting simulation studies on a benchmark CSTR system [W. C. Li and L. T. Biegler, Process control strategies for constrained nonlinear systems, Ind. Eng. Chem. Res. 27, 142 (1988)] which exhibits input multiplicity and change in the sign of steady state gain in the operating region.
For the entire collection see [Zbl 1116.93009].


93C10 Nonlinear systems in control theory
93B30 System identification
93E12 Identification in stochastic control theory
93C57 Sampled-data control/observation systems


Zbl 0874.93034
Full Text: DOI