An extended Marquardt-type procedure for fitting error-in-variables models

The paper presents a simple derivation of the method of fitting nonlinear algebraic models where all variables are subject to error and improves the numerical efficiency of the algorithm. Including a known procedure for equilibrating balance equations and factorizing the weighting matrix, the classi...

Full description

Saved in:
Bibliographic Details
Published inComputers & chemical engineering Vol. 11; no. 1; pp. 37 - 43
Main Authors Valkó, P., Vajda, S.
Format Journal Article
LanguageEnglish
Published Oxford Elsevier Ltd 1987
Elsevier
Subjects
Applications of mathematics to chemical engineering. Modeling. Simulation. Optimization
Applied sciences
Chemical engineering
Exact sciences and technology
Non linear equation
Algebraic equation
Error
Parameter estimation
Mathematical model
Chemical engineering
Online AccessGet full text

Cover

More Information
Summary:The paper presents a simple derivation of the method of fitting nonlinear algebraic models where all variables are subject to error and improves the numerical efficiency of the algorithm. Including a known procedure for equilibrating balance equations and factorizing the weighting matrix, the classical Gauss-Marquardt method of estimating parameters in nonlinear models is shown to handle also the error-in-variables model, thereby extending the efficiency and robustness of Marquardt's compromise to this slightly more involved case.
ISSN:0098-1354
1873-4375
DOI:10.1016/0098-1354(87)80004-2