System identification using an over-parametrized model class-improving the optimization algorithm
Authors:
- T. McKelveyTMT. McKelvey
- A. HelmerssonAHA. Helmersson
- 1 January 1999
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 3, 2984-2989
Abstract
The use of an over-parametrized state-space model for system identification has some clear advantages: A single model structure covers the entire class of multivariable systems up to a given order. The over-parametrization also leads to the possibility to choose a numerically stable parametrization. During the parametric optimization the gradient calculations constitute the main computational part of the algorithm. Consequently using more than theminimal number of parameters requiredslows down thealgorithm. However, we show that for any chosen (over)-parametrization it is possible to reduce the gradientcalculations to the minimal amount by constructing the parameter subspace which is orthonormal to the tangent space of the manifold representing equivalent modelsKeywords
This publication has 7 references indexed in Scilit:
- An LFT approach to parameter estimation1997
- State-space parametrizations of multivariable linear systems using tridiagonal matrix forms1996
- Balanced realizations: canonical form, parametrization, model reductionInternational Journal of Control, 1987
- On-line structure selection for multivariable state-space modelsAutomatica, 1982
- Canonical structures in the identification of multivariable systemsAutomatica, 1975
- Computational aspects of maximum likelihood estimation and reduction in sensitivity function calculationsIEEE Transactions on Automatic Control, 1974
- Canonical forms for linear multivariable systemsIEEE Transactions on Automatic Control, 1967