Surrogate model for Bayesian optimal experimental design in chromatography

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Rojo-Garcia, Jose Rodrigo, Haario, Heikki, Helin, Tapio, Sainio, Tuomo
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866914966293970944
author Rojo-Garcia, Jose Rodrigo
Haario, Heikki
Helin, Tapio
Sainio, Tuomo
author_facet Rojo-Garcia, Jose Rodrigo
Haario, Heikki
Helin, Tapio
Sainio, Tuomo
contents We applied Bayesian Optimal Experimental Design (OED) in the estimation of parameters involved in the Equilibrium Dispersive Model for chromatography with two components with the Langmuir adsorption isotherm. The coefficients estimated were Henry's coefficients, the total absorption capacity and the number of theoretical plates, while the design variables were the injection time and the initial concentration. The Bayesian OED algorithm is based on nested Monte Carlo estimation, which becomes computationally challenging due to the simulation time of the PDE involved in the dispersive model. This complication was relaxed by introducing a surrogate model based on Piecewise Sparse Linear Interpolation. Using the surrogate model instead the original reduces significantly the simulation time and it approximates the solution of the PDE with high degree of accuracy. The estimation of the parameters over strategical design points provided by OED reduces the uncertainty in the estimation of parameters. Additionally, the Bayesian OED methodology indicates no improvements when increasing the number of measurements in temporal nodes above a threshold value.
format Preprint
id arxiv_https___arxiv_org_abs_2406_19835
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Surrogate model for Bayesian optimal experimental design in chromatography
Rojo-Garcia, Jose Rodrigo
Haario, Heikki
Helin, Tapio
Sainio, Tuomo
Applications
Numerical Analysis
62Kxx, 62Pxx, 62F15, 35R30
We applied Bayesian Optimal Experimental Design (OED) in the estimation of parameters involved in the Equilibrium Dispersive Model for chromatography with two components with the Langmuir adsorption isotherm. The coefficients estimated were Henry's coefficients, the total absorption capacity and the number of theoretical plates, while the design variables were the injection time and the initial concentration. The Bayesian OED algorithm is based on nested Monte Carlo estimation, which becomes computationally challenging due to the simulation time of the PDE involved in the dispersive model. This complication was relaxed by introducing a surrogate model based on Piecewise Sparse Linear Interpolation. Using the surrogate model instead the original reduces significantly the simulation time and it approximates the solution of the PDE with high degree of accuracy. The estimation of the parameters over strategical design points provided by OED reduces the uncertainty in the estimation of parameters. Additionally, the Bayesian OED methodology indicates no improvements when increasing the number of measurements in temporal nodes above a threshold value.
title Surrogate model for Bayesian optimal experimental design in chromatography
topic Applications
Numerical Analysis
62Kxx, 62Pxx, 62F15, 35R30
url https://arxiv.org/abs/2406.19835