Extension of static non-linear doe identification algorithms to dynamic systems

Authors:
Markus Hirsch;Thomas E. Passenbrunner
Affiliations:
Institute for Design and Control of Mechatronical Systems, Johannes Kepler University, Linz, Austria;Institute for Design and Control of Mechatronical Systems, Johannes Kepler University, Linz, Austria
Venue:
EUROCAST'11 Proceedings of the 13th international conference on Computer Aided Systems Theory - Volume Part II
Year:
2011

Citing 3
Cited 0

Nonlinear black-box modeling in system identification: a unified overview

Automatica (Journal of IFAC) - Special issue on trends in system identification
System identification (2nd ed.): theory for the user

System identification (2nd ed.): theory for the user
Survey paper: Optimal experimental design and some related control problems

Automatica (Journal of IFAC)

Quantified Score

Hi-index	0.00

Visualization

Abstract

The excitation of a system is a crucial task in system identification. If the data was not captured with persistent excitation, the parameter estimation degrades and the parameter variance increases. Application of Design of Experiment (DoE) algorithms helps to overcome these problems. This work presents the extension of the well known Wynn-Algorithm for non-linear but static systems to non-linear dynamic systems. Instead of optimizing the input only for the next sampling instant, an input sequence is optimized over a receding horizon. The developed algorithm is used to identify the model of a hydrodynamic brake used on a combustion engine test bench as load machine.