Skip to main content

LUP Student Papers

LUND UNIVERSITY LIBRARIES

Duala regulatorer

Abramowicz, Henrik and Stymne, Karl-Johan (1975) In MSc Theses
Department of Automatic Control
Abstract
Our task has been to compare the ability of three different regulators to control some single-input single-output systems with unknown parameters. The regulators were a minimal variance regulator, the Self-TUning REgulator STURE and an active suboptimally dual regulator. No simulations with the minimal variance regulator are shown since it turned out that it performs as STURE on systems with constant parameters and it does not work at all on systems with time-varying parameters (one time-delay). On systems with constant parameters STURE and the active sub-optimal regulator gave almost identical losses in stationarity (=the loss with a minimal variance regulator for known parameters), but it seems as if STURE converges quicker. These two... (More)
Our task has been to compare the ability of three different regulators to control some single-input single-output systems with unknown parameters. The regulators were a minimal variance regulator, the Self-TUning REgulator STURE and an active suboptimally dual regulator. No simulations with the minimal variance regulator are shown since it turned out that it performs as STURE on systems with constant parameters and it does not work at all on systems with time-varying parameters (one time-delay). On systems with constant parameters STURE and the active sub-optimal regulator gave almost identical losses in stationarity (=the loss with a minimal variance regulator for known parameters), but it seems as if STURE converges quicker. These two regulators gave approximately the same loss also on systems with time-varying parameters except when the b-parameter, i.e. the gain, could change sign. In these cases the system became unstable when controlled by STURE, while the active suboptimal regulator managed to control. Our results seem to indicate that it is only if the gain may change sign that it pays to use a more complex regulator than STURE. (Less)
Please use this url to cite or link to this publication:
author
Abramowicz, Henrik and Stymne, Karl-Johan
supervisor
organization
year
type
H3 - Professional qualifications (4 Years - )
subject
publication/series
MSc Theses
report number
TFRT-5159
ISSN
0346-5500
language
Swedish
id
8850357
date added to LUP
2016-03-29 10:23:50
date last changed
2016-03-29 10:23:50
@misc{8850357,
  abstract     = {{Our task has been to compare the ability of three different regulators to control some single-input single-output systems with unknown parameters. The regulators were a minimal variance regulator, the Self-TUning REgulator STURE and an active suboptimally dual regulator. No simulations with the minimal variance regulator are shown since it turned out that it performs as STURE on systems with constant parameters and it does not work at all on systems with time-varying parameters (one time-delay). On systems with constant parameters STURE and the active sub-optimal regulator gave almost identical losses in stationarity (=the loss with a minimal variance regulator for known parameters), but it seems as if STURE converges quicker. These two regulators gave approximately the same loss also on systems with time-varying parameters except when the b-parameter, i.e. the gain, could change sign. In these cases the system became unstable when controlled by STURE, while the active suboptimal regulator managed to control. Our results seem to indicate that it is only if the gain may change sign that it pays to use a more complex regulator than STURE.}},
  author       = {{Abramowicz, Henrik and Stymne, Karl-Johan}},
  issn         = {{0346-5500}},
  language     = {{swe}},
  note         = {{Student Paper}},
  series       = {{MSc Theses}},
  title        = {{Duala regulatorer}},
  year         = {{1975}},
}