Optimal experiment design for regression polynomial models identification

DOI10.1080/00207170210159131MaRDI QIDQ4800365zbMATH OpenOpenAlexFDO

Authors G. Belforte, Paolo Gay

Publication date 9 November 2003

Published in International Journal of Control (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1080/00207170210159131

zbMATH Keywords

identificationoptimal input sequencesoptimal interpolationregression polynomial modelsstatic nonlinear blockVandermonde parametrization

Mathematics Subject Classification ID

System identification (93B30) Nonlinear systems in control theory (93C10)

Review text

The paper deals with the identification of static nonlinear blocks based on the optimal interpolation problem for polynomials described in the Vandermonde parametrization, in which the set of optimal input values is derived when the input varies on a given finite interval and the cost functions to be minimized are the worst case parameter uncertainties. Similar results for representing polynomials with different bases, still preserving the optimal set of input levels derived from the Vandermonde parametrization, are presented too. Finally some numerical examples show the effectiveness of the optimal input sequences, especially when identifying some block described dynamic models that structurally include static nonlinearities, where the confidence of the estimates can add up to a factor of several hundred with respect to the case of random generated inputs.

Recommendations

• Optimal designs for identifying the degree of a polynomial regression
• scientific article; zbMATH DE number 4060570
• scientific article; zbMATH DE number 4072184
• Optimum experimental design in nonlinear regression
• \(E\)-optimal designs for polynomial regression
• scientific article; zbMATH DE number 788258

Cites work

• A new method for the identification of Hammerstein model
• A nonparametric polynomial identification algorithm for the Hammerstein system
• Dynamic system identification. Experiment design and data analysis
• scientific article; zbMATH DE number 3945143 (Why is no real title available?)
• Optimal input design for worst-case system identification in \(l_ 1/l_ 2/l_ \infty\)
• Optimal sampling schedule for parameter estimation of linear models with unknown but bounded measurement errors
• Optimal worst case estimation for LPV--FIR models with bounded errors
• Some new properties of Chebyshev polynomials
• Structure identification of nonlinear dynamic systems - A survey on input/output approaches

Cited in

(4)

• Experiment planning and identification in a quasilinear regression problem
• Optimal worst case estimation for LPV--FIR models with bounded errors
• System identification of complex and structured systems
• scientific article; zbMATH DE number 2092126 (Why is no real title available?)

This page was built for publication: Optimal experiment design for regression polynomial models identification

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q4800365)