UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

An optimal experimental design strategy for improving parameter estimation in stochastic models

Huang, Chunbing; Cattani, Federica; Galvanin, Federico; (2023) An optimal experimental design strategy for improving parameter estimation in stochastic models. Computers and Chemical Engineering , 170 , Article 108133. 10.1016/j.compchemeng.2023.108133. Green open access

[thumbnail of Kandt_Who is returning to public transport for non-work trips after COVID-19_VoR.pdf]
Preview
Text
Kandt_Who is returning to public transport for non-work trips after COVID-19_VoR.pdf

Download (5MB) | Preview

Abstract

Systematic model-based design of experiment is essential to maximise the information from an experimental campaign. This technique is even more important to design experiments in systems described by stochastic models where the information quantity is characterised by intrinsic uncertainty, which has a significant impact on the experimental design for yielding informative data for precisely estimating model parameters. In this work, a new method for stochastic model-based design of experiments (SMBDoE) is presented to simultaneously identify the optimal operating conditions and the allocation of sampling points in time. The optimal experiment is identified by two sampling strategies selecting sampling intervals based on the average and the uncertainty of Fisher information. Seed coating is used as a case study to illustrate the feasibility of the method in identifying optimal coating conditions and sampling strategy in an industrial application.

Type: Article
Title: An optimal experimental design strategy for improving parameter estimation in stochastic models
Open access status: An open access version is available from UCL Discovery
DOI: 10.1016/j.compchemeng.2023.108133
Publisher version: https://doi.org/10.1016/j.compchemeng.2023.108133
Language: English
Additional information: © 2023 The Author(s). Published by Elsevier Ltd. under a Creative Commons license (http://creativecommons.org/licenses/by/4.0/).
Keywords: Stochastic model-based design of experiments, stochastic simulation, seed coating system, sampling strategy
UCL classification: UCL
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Chemical Engineering
URI: https://discovery.ucl.ac.uk/id/eprint/10162690
Downloads since deposit
18Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item