Optimal input design for reduction of parameter correlations

Wang, Ke and Yue, Hong and Yu, Hui (2018) Optimal input design for reduction of parameter correlations. In: The 24th International Conference on Automation and Computing (ICAC'18), 2018-09-06 - 2018-09-07, Newcastle University. (https://doi.org/10.23919/IConAC.2018.8749035)

[thumbnail of Wang-etal-ICAC-2018-Optimal-input-design-for-reduction-of-parameter-correlations]
Preview
 Text. Filename: Wang_etal_ICAC_2018_Optimal_input_design_for_reduction_of_parameter_correlations.pdf
Accepted Author Manuscript
License: Strathprints license 1.0
Download (1MB)| Preview

Abstract

An new scalarisation criterion is proposed for optimal experiment design (OED) of input intensity so as to obtain the most informative experimental data for parameter estimation with reduced parameter correlations. This criterion is a linear combination of logarithm function of the A-optimality and the modified E (ME)-optimality. It can be used to improve the estimation quality from the A-optimal design, and to reduce parameter correlations from the MEoptimal design. The proposed algorithm has been examined through simulation study of an enzyme reaction system model. The results are compared with A-optimal design, MEoptimal design, and other designs with a focus on reducing parameter correlations such as the C- and the CE- designs.

CORE (COnnecting REpositories)