Uncertainty quantification of property models: Methodology and its application to CO₂‐loaded aqueous MEA solutions

Abstract

in Wiley Online Library (wileyonlinelibrary.com) Uncertainties in property models can significantly affect the results obtained from process simulations. If these uncertainties are not quantified, optimal plant designs based on such models can be misleading. With this incentive, a systematic, generalized uncertainty quantification (UQ) methodology for property models is developed. Starting with prior beliefs about parametric uncertainties, a Bayesian method is used to derive informed posteriors using the experimental data. To reduce the computational expense, surrogate response surface models are developed. For downselecting the parameter space, a sensitivity matrix-based approach is developed. The methodology is then deployed to the property models for an MEA-CO 2 -H 2 O system. The UQ analysis is found to provide interesting information about uncertainties in the parameter space. The sensitivity matrix approach is also found to be a valuable tool for reducing computational expense. Finally, the effect of the estimated parametric uncertainty on CO 2 absorption and monoethanolamine (MEA) regeneration is analyzed.

Impact of Uncertainties on the Design and Cost of CCS From a Waste-to-Energy Plant

“…While MEA-based CO 2 capture is not the most energy efficient CO 2 capture technology option, it is one of the most mature and well-characterized CO 2 capture technologies. It thus presents limited uncertainties (Morgan et al, 2015;Kuncheekanna et al, 2019).…”

Toward transformational carbon capture systems

“…Other times, inadequate data are used during the development of the model because it is difficult to measure the properties under those conditions. Ultimately, this can be a critical oversight causing the overall model to perform poorly if the submodel is extrapolating well beyond the range of conditions under which it was developed …”

Physical Property Modeling of Solvent-Based Carbon Capture Processes with Uncertainty Quantification and Validation with Pilot Plant Data

“…The goal of the physical property modeling is to create a framework that is widely applicable to property modeling for any system, including uncertainty quantification. The contributions of this portion of the work are included in two manuscripts, for standalone property models such as viscosity, density, and molar volume of the MEA-CO 2 -H 2 O system (Morgan et al, 2015) and the thermodynamic framework of this system. (Morgan et al, 2017a).…”