A Modular Approach for Surrogate Modeling of Flowsheets. Issue 12 (23rd September 2021)

• Record Type: Journal Article
• Authors: Bubel, Martin; Ludl, Patrick Otto; Seidel, Tobias; Asprion, Norbert; Bortz, Michael
• Other Names: Bortz Michael guestEditor.; Dadhe Kai guestEditor.; Mitsos Alexander guestEditor.
• Is Part Of: Chemie Ingenieur Technik. Volume 93:Issue 12(2021)
• Page Start: 1987
• Page End: 1997
• Publication Date: 2021-09-23
• Subjects: Flowsheet simulation -- Process simulation -- Surrogate modeling; Chemical engineering -- Patents -- Periodicals; Chemical engineering -- Periodicals; Chemical industry -- Periodicals; Chemistry, Technical -- Periodicals; 660.05
• DOI: 10.1002/cite.202100077
• Languages: English
• ISSNs: 0009-286X

Abstract

Abstract: Flowsheet simulation, an important building block in chemical process development and design, generally requires the solution of a nonlinear system of equations. Such a simulation can fail if either no solution exists for the chosen specifications or if the initial values for the solver are chosen unfavorably. In this work, to overcome such convergence failures, AI‐based surrogate models are trained for unit operations of a flowsheet and then interconnected. For a pressure swing distillation with recycle, it is shown that such an interconnection of surrogate models yields more accurate results compared to a surrogate model for the whole flowsheet at once, and that adequate starting points for the flowsheet simulation can be obtained from the interconnected surrogates. Abstract : AI‐based learning methods are applied to unit operations of a flowsheet and then interconnected. The interconnection of surrogate models is shown to be favorable compared to a surrogate model for the whole flowsheet. Also, adequate starting points for a commercial simulator are obtained from the interconnected surrogate models.