Advanced

Robust constrained Model Predictive Control of fast electromechanical systems

Blanchini, Franco; Casagrande, Daniele; Giordano, Giulia LU and Viaro, Umberto (2016) In Journal of the Franklin Institute 353(9). p.2087-2103
Abstract

A major drawback hinders the application of Model Predictive Control (MPC) to the regulation of electromechanical systems or, more generally, systems with fast dynamics: the time needed for the online computation of the control is often too long with respect to the sampling time. This paper shows how this problem can be overcome by suitably implementing the MPC technique. The main idea is to compute the control law using the discrete-time Euler Auxiliary System (EAS) associated with the continuous-time plant, and apply the control obtained for the discrete-time system to the continuous-time system. In this way the implementation sampling time can be much smaller than the EAS time parameter, which leads to significant savings in... (More)

A major drawback hinders the application of Model Predictive Control (MPC) to the regulation of electromechanical systems or, more generally, systems with fast dynamics: the time needed for the online computation of the control is often too long with respect to the sampling time. This paper shows how this problem can be overcome by suitably implementing the MPC technique. The main idea is to compute the control law using the discrete-time Euler Auxiliary System (EAS) associated with the continuous-time plant, and apply the control obtained for the discrete-time system to the continuous-time system. In this way the implementation sampling time can be much smaller than the EAS time parameter, which leads to significant savings in computation time. Theoretical results guarantee stabilisation, constraint satisfaction and robustness of such a control strategy, which is applied to the control of an electric drive and a cart-pendulum system.

(Less)
Please use this url to cite or link to this publication:
author
publishing date
type
Contribution to journal
publication status
published
subject
in
Journal of the Franklin Institute
volume
353
issue
9
pages
17 pages
publisher
Elsevier
external identifiers
  • Scopus:84963538298
ISSN
0016-0032
DOI
10.1016/j.jfranklin.2016.03.009
language
English
LU publication?
no
id
49bb6eb2-1fff-4ca8-b0a5-e415e91b1d7a
date added to LUP
2016-07-06 15:15:09
date last changed
2016-07-11 09:44:37
@misc{49bb6eb2-1fff-4ca8-b0a5-e415e91b1d7a,
  abstract     = {<p>A major drawback hinders the application of Model Predictive Control (MPC) to the regulation of electromechanical systems or, more generally, systems with fast dynamics: the time needed for the online computation of the control is often too long with respect to the sampling time. This paper shows how this problem can be overcome by suitably implementing the MPC technique. The main idea is to compute the control law using the discrete-time Euler Auxiliary System (EAS) associated with the continuous-time plant, and apply the control obtained for the discrete-time system to the continuous-time system. In this way the implementation sampling time can be much smaller than the EAS time parameter, which leads to significant savings in computation time. Theoretical results guarantee stabilisation, constraint satisfaction and robustness of such a control strategy, which is applied to the control of an electric drive and a cart-pendulum system.</p>},
  author       = {Blanchini, Franco and Casagrande, Daniele and Giordano, Giulia and Viaro, Umberto},
  issn         = {0016-0032},
  language     = {eng},
  month        = {06},
  number       = {9},
  pages        = {2087--2103},
  publisher    = {ARRAY(0x9979a48)},
  series       = {Journal of the Franklin Institute},
  title        = {Robust constrained Model Predictive Control of fast electromechanical systems},
  url          = {http://dx.doi.org/10.1016/j.jfranklin.2016.03.009},
  volume       = {353},
  year         = {2016},
}