Sensitivity analysis of adsorption isotherms subject to measurement noise in the data

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The analysis and design of many environmental and separation processes rely on the availability of accurate adsorption isotherms. These isotherms are estimated from measurements of the adsorption process variables. Unfortunately, these variables are usually contaminated with errors that affect the accuracy of their estimated parameters. Therefore, one objective of this work was to study the effect of measurement noise in the variables on the estimation accuracy of the Langmuir isotherm. In fact, Langmuir has three linearized forms, so it was sought to determine which out of these three forms would provide the most accurate estimated parameters. Another objective of this work was to estimate the isotherm parameters using the nonlinear Langmuir form using nonlinear optimization, and to compare the accuracy of the estimated parameters to the ones obtained using the most accurate linearized form. A third objective was to study the effect of measurement noise level on the accuracy of the Langmuir isotherm. As a result of this study, the following was found. One of the three linearized Langmuir forms provided the most accurate estimates. In fact, its accuracy was even comparable to that obtained by nonlinear optimization using the nonlinear isotherm. In addition, the estimation accuracy was more sensitive to the magnitude of the affinity constant than to the maximum amount of adsorbate in adsorbent; larger values of affinity constant result in higher estimation accuracy of both model parameters. Finally, it was confirmed that the higher the noise content in the variables, the larger the uncertainty of their estimation.

Original languageEnglish
Title of host publicationAIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings
PublisherAmerican Institute of Chemical Engineers
ISBN (Print)9780816910502
Publication statusPublished - 1 Jan 2008

Publication series

NameAIChE Annual Meeting, Conference Proceedings

Fingerprint

Adsorption isotherms
Sensitivity analysis
Isotherms
Adsorbates
Adsorbents
Availability
Adsorption

Keywords

  • Estimation
  • Isotherms
  • Langmuir
  • Measurement noise
  • Sensitivity analysis

ASJC Scopus subject areas

  • Chemical Engineering(all)
  • Chemistry(all)

Cite this

Farhat, K. K., Nounou, M., & Abdel-Wahab, A. (2008). Sensitivity analysis of adsorption isotherms subject to measurement noise in the data. In AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings (AIChE Annual Meeting, Conference Proceedings). American Institute of Chemical Engineers.

Sensitivity analysis of adsorption isotherms subject to measurement noise in the data. / Farhat, Karim K.; Nounou, Mohamed; Abdel-Wahab, Ahmed.

AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings. American Institute of Chemical Engineers, 2008. (AIChE Annual Meeting, Conference Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Farhat, KK, Nounou, M & Abdel-Wahab, A 2008, Sensitivity analysis of adsorption isotherms subject to measurement noise in the data. in AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings. AIChE Annual Meeting, Conference Proceedings, American Institute of Chemical Engineers.
Farhat KK, Nounou M, Abdel-Wahab A. Sensitivity analysis of adsorption isotherms subject to measurement noise in the data. In AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings. American Institute of Chemical Engineers. 2008. (AIChE Annual Meeting, Conference Proceedings).
Farhat, Karim K. ; Nounou, Mohamed ; Abdel-Wahab, Ahmed. / Sensitivity analysis of adsorption isotherms subject to measurement noise in the data. AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings. American Institute of Chemical Engineers, 2008. (AIChE Annual Meeting, Conference Proceedings).
@inproceedings{94ae0a0ed75c4e7d8cb9b789ca7ece79,
title = "Sensitivity analysis of adsorption isotherms subject to measurement noise in the data",
abstract = "The analysis and design of many environmental and separation processes rely on the availability of accurate adsorption isotherms. These isotherms are estimated from measurements of the adsorption process variables. Unfortunately, these variables are usually contaminated with errors that affect the accuracy of their estimated parameters. Therefore, one objective of this work was to study the effect of measurement noise in the variables on the estimation accuracy of the Langmuir isotherm. In fact, Langmuir has three linearized forms, so it was sought to determine which out of these three forms would provide the most accurate estimated parameters. Another objective of this work was to estimate the isotherm parameters using the nonlinear Langmuir form using nonlinear optimization, and to compare the accuracy of the estimated parameters to the ones obtained using the most accurate linearized form. A third objective was to study the effect of measurement noise level on the accuracy of the Langmuir isotherm. As a result of this study, the following was found. One of the three linearized Langmuir forms provided the most accurate estimates. In fact, its accuracy was even comparable to that obtained by nonlinear optimization using the nonlinear isotherm. In addition, the estimation accuracy was more sensitive to the magnitude of the affinity constant than to the maximum amount of adsorbate in adsorbent; larger values of affinity constant result in higher estimation accuracy of both model parameters. Finally, it was confirmed that the higher the noise content in the variables, the larger the uncertainty of their estimation.",
keywords = "Estimation, Isotherms, Langmuir, Measurement noise, Sensitivity analysis",
author = "Farhat, {Karim K.} and Mohamed Nounou and Ahmed Abdel-Wahab",
year = "2008",
month = "1",
day = "1",
language = "English",
isbn = "9780816910502",
series = "AIChE Annual Meeting, Conference Proceedings",
publisher = "American Institute of Chemical Engineers",
booktitle = "AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings",
address = "United States",

}

TY - GEN

T1 - Sensitivity analysis of adsorption isotherms subject to measurement noise in the data

AU - Farhat, Karim K.

AU - Nounou, Mohamed

AU - Abdel-Wahab, Ahmed

PY - 2008/1/1

Y1 - 2008/1/1

N2 - The analysis and design of many environmental and separation processes rely on the availability of accurate adsorption isotherms. These isotherms are estimated from measurements of the adsorption process variables. Unfortunately, these variables are usually contaminated with errors that affect the accuracy of their estimated parameters. Therefore, one objective of this work was to study the effect of measurement noise in the variables on the estimation accuracy of the Langmuir isotherm. In fact, Langmuir has three linearized forms, so it was sought to determine which out of these three forms would provide the most accurate estimated parameters. Another objective of this work was to estimate the isotherm parameters using the nonlinear Langmuir form using nonlinear optimization, and to compare the accuracy of the estimated parameters to the ones obtained using the most accurate linearized form. A third objective was to study the effect of measurement noise level on the accuracy of the Langmuir isotherm. As a result of this study, the following was found. One of the three linearized Langmuir forms provided the most accurate estimates. In fact, its accuracy was even comparable to that obtained by nonlinear optimization using the nonlinear isotherm. In addition, the estimation accuracy was more sensitive to the magnitude of the affinity constant than to the maximum amount of adsorbate in adsorbent; larger values of affinity constant result in higher estimation accuracy of both model parameters. Finally, it was confirmed that the higher the noise content in the variables, the larger the uncertainty of their estimation.

AB - The analysis and design of many environmental and separation processes rely on the availability of accurate adsorption isotherms. These isotherms are estimated from measurements of the adsorption process variables. Unfortunately, these variables are usually contaminated with errors that affect the accuracy of their estimated parameters. Therefore, one objective of this work was to study the effect of measurement noise in the variables on the estimation accuracy of the Langmuir isotherm. In fact, Langmuir has three linearized forms, so it was sought to determine which out of these three forms would provide the most accurate estimated parameters. Another objective of this work was to estimate the isotherm parameters using the nonlinear Langmuir form using nonlinear optimization, and to compare the accuracy of the estimated parameters to the ones obtained using the most accurate linearized form. A third objective was to study the effect of measurement noise level on the accuracy of the Langmuir isotherm. As a result of this study, the following was found. One of the three linearized Langmuir forms provided the most accurate estimates. In fact, its accuracy was even comparable to that obtained by nonlinear optimization using the nonlinear isotherm. In addition, the estimation accuracy was more sensitive to the magnitude of the affinity constant than to the maximum amount of adsorbate in adsorbent; larger values of affinity constant result in higher estimation accuracy of both model parameters. Finally, it was confirmed that the higher the noise content in the variables, the larger the uncertainty of their estimation.

KW - Estimation

KW - Isotherms

KW - Langmuir

KW - Measurement noise

KW - Sensitivity analysis

UR - http://www.scopus.com/inward/record.url?scp=85069300701&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85069300701&partnerID=8YFLogxK

M3 - Conference contribution

AN - SCOPUS:85069300701

SN - 9780816910502

T3 - AIChE Annual Meeting, Conference Proceedings

BT - AIChE100 - 2008 AIChE Annual Meeting, Conference Proceedings

PB - American Institute of Chemical Engineers

ER -