Recursive identification of time-varying systems | Jens Ejnar Parkum
| | Abstract | The present thesis is concerned with recursive identification of time-varying systems. Methods designed for estimating slowly drifting system parameters are studied. A general algorithm containing most existing methods as special cases is formulated, and a unified treatment of its convergence properties in an ideal deterministic environment is presented. An algorithm which discounts information non-uniformly and in accordance with the distribution of the information flow is developed and examined. Finally, the discussed techniques are applied to a practical problem with adaptive control of a non-linear process. | | Type | Ph.D. thesis [Academic thesis] | | Year | 1992 | | Publisher | Informatics and Mathematical Modelling, Technical University of Denmark, DTU | | Address | Richard Petersens Plads, Building 321, DK-2800 Kgs. Lyngby | | Series | IMSOR-PHD-1992-57 | | Note | Supervised by Niels K. Poulsen, and Henrik Spliid, DTU Informatics, http://www.imm.dtu.dk | | Electronic version(s) | [pdf] | | BibTeX data | [bibtex] | | IMM Group(s) | Mathematical Statistics |
|