Projection-based identification algorithm for grey-box continuous-time models

In this paper, a new identification method for continuous-time models, which can handle various grey-box structures and has strong robustness, is presented. The proposed method is based on an incremental model update scheme and the projection onto the subspace which reflects the model structure. By...

Full description

Saved in:
Bibliographic Details
Published inSystems & control letters Vol. 62; no. 11; pp. 1090 - 1097
Main Authors Maruta, Ichiro, Sugie, Toshiharu
Format Journal Article
LanguageEnglish
Published Elsevier B.V 01.11.2013
Subjects
Algorithms
Continuous-time system identification
Control systems
Grey-box model
Handles
Mathematical models
Optimization
Projection
Robustness
Time-delay system
Versatility
Time-delay system
Continuous-time system identification
Grey-box model
Online AccessGet full text

Cover

More Information
Summary:In this paper, a new identification method for continuous-time models, which can handle various grey-box structures and has strong robustness, is presented. The proposed method is based on an incremental model update scheme and the projection onto the subspace which reflects the model structure. By utilising these schemes, robustness of other continuous-time system identification methods and versatility of generic optimisation algorithms can be integrated into the proposed method. The effectiveness of the proposed method is demonstrated through numerical examples related to a grey-box model in closed-loop system and systems with unknown time-delay.
Bibliography:ObjectType-Article-2
SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 23
ISSN:0167-6911
1872-7956
DOI:10.1016/j.sysconle.2013.08.006