Data quality and assessment, validation methods and error propagation through the simulation software: Report from the Round-Table Discussion at the 10th World Congress of Chemical Engineering in Barcelona (October 1–5, 2017)

P. Mathias (1) , A. Soto (2) , L. Fele-Zilnik (3) , J.-C de Hemptinne (4) , A. Bazyleva (5) , J. Abildskov (6)

1 Fluor corporation
2 USC - Universidade de Santiago de Compostela [Spain]
3 National Institute of Chemistry
4 IFPEN - IFP Energies nouvelles
5 NIST - National Institute of Standards and Technology [Gaithersburg]
6 DTU - Danmarks Tekniske Universitet = Technical University of Denmark

The issues of data quality and propagation of data uncertainties into process design and plant specifications are of great current interest. Hence, two Working Parties of the European Federation of Chemical Engineers (EFCE) organized a Round Table Discussion on the topic, as part of the World Congress of Chemical Engineering (WCCE10) in Barcelona, in October 2017. The discussion was guided by industrial and academic experts, with the audience as a key part of the discussion, trying to find some answers in three areas: Data acquisition and evaluation of experimental uncertainties, tools for data reconciliation to improve their quality, and impact of data uncertainties on the process at the end. Several concrete stories are presented that demonstrate the importance of considering data quality and all possible contributions to the uncertainty of chemical process design. Difficulties associated with data quality are discussed at various levels: (1) the experimentalists (measurement issues, evaluation of uncertainties, use of consistency analysis tools); (2) model developers (capture of adequate physics, parameter regression strategies, uncertainty propagation), (3) vendors of process simulation software, and (4) process engineers (who are responsible at the end). Paths for improvements were proposed through better and more efficient communication among different participants, as well as through education.

Keywords: error propagation; data quality and assessment; data reconciliation; uncertainties; process simulation

Domains: Chemical Sciences

P. Mathias, A. Soto, L. Fele-Zilnik, J.-C de Hemptinne, A. Bazyleva, et al.. Data quality and assessment, validation methods and error propagation through the simulation software: Report from the Round-Table Discussion at the 10th World Congress of Chemical Engineering in Barcelona (October 1–5, 2017). Chemical Engineering Research and Design, 2018, 137, pp.A1-A8. ⟨10.1016/j.cherd.2018.08.010⟩. ⟨hal-01948470⟩