Generalised Disjunctive Programming for Process Synthesis

  • Automating process synthesis presents a formidable challenge in chemical engineering. Par-ticularly challenging is the development of frameworks that are both general and accurate, while remaining computationally tractable. To achieve generality, a building block-based modelling approach was proposed in previous contributions by Kuhlmann and Skiborowski and Krone et al.. This model formulation incorporates Phenomena-based Building Blocks (PBBs), capable of depicting a wide array of separation processes. To maximize accu-racy, the PBBs are interfaced with CAPE-OPEN thermodynamics, allowing for detailed ther-modynamic models within the process synthesis problem. However, the pursuit of gener-ality and accuracy introduces increased model complexity and poses the risk of combinatori-al explosion. To address this and enhance tractability, developed a structural screening method that forbids superstructures leading to infeasible configurations. These combined innovations allow for general,Automating process synthesis presents a formidable challenge in chemical engineering. Par-ticularly challenging is the development of frameworks that are both general and accurate, while remaining computationally tractable. To achieve generality, a building block-based modelling approach was proposed in previous contributions by Kuhlmann and Skiborowski and Krone et al.. This model formulation incorporates Phenomena-based Building Blocks (PBBs), capable of depicting a wide array of separation processes. To maximize accu-racy, the PBBs are interfaced with CAPE-OPEN thermodynamics, allowing for detailed ther-modynamic models within the process synthesis problem. However, the pursuit of gener-ality and accuracy introduces increased model complexity and poses the risk of combinatori-al explosion. To address this and enhance tractability, developed a structural screening method that forbids superstructures leading to infeasible configurations. These combined innovations allow for general, accurate, and tractable superstructures. To further increase the solvable problem size, we propose an advanced optimization frame-work, leveraging generalized disjunctive programming (GDP). It allows for multiple im-provements over existing MINLP formulations, aiming at improving feasibility and solution time. This is achieved by deactivation of unused model equations during the solution proce-dure. Additionally, Grossmann showed that a disjunctive branch-and-bound algorithm can be postulated. This provides tighter bounds for linear problems than those obtained through reformulations used in conventional MINLP solvers, reducing the required solution time. Building on these insights, it is of interest whether these findings extend to nonlinear sys-tems. To investigate this, we developed a MathML/XML-based automatic code generation tool inside MOSAICmodeling, which formulates complex nonlinear GDP and exports them to conventional optimization environments (Pyomo, GAMS etc.). These are then coupled with structural screening methods and solved using out-of-the-box functionalities for GDP solution. To validate the proposed approach, a case study is conducted involving two PBBs, previously published by Krone et al.. The study compares the performance of the GDP-based optimization framework against conventional MINLP approaches. Preliminary results suggest that the GDP-based framework offers computational advantages over conventional MINLP formulations. The full paper will present detailed comparisons, offering insights into the practical applicability and benefits of GDP.zeige mehrzeige weniger

Volltext Dateien herunterladen

  • 2025-07-06-Poster-ESCAPE35-Scheffold-final - A0.pdf
    eng

Metadaten exportieren

Weitere Dienste

Suche bei Google Scholar Anzahl der Zugriffe auf dieses Dokument
Metadaten
Autor*innen:Lukas ScheffoldORCiD, Erik EscheORCiD
Dokumenttyp:Posterpräsentation
Veröffentlichungsform:Präsentation
Sprache:Englisch
Jahr der Erstveröffentlichung:2025
Organisationseinheit der BAM:2 Prozess- und Anlagensicherheit
2 Prozess- und Anlagensicherheit / 2.2 Prozesssimulation
DDC-Klassifikation:Technik, Medizin, angewandte Wissenschaften / Chemische Verfahrenstechnik / Chemische Verfahrenstechnik
Freie Schlagwörter:Distillation Column; Generalized Disjunctive Programming; Process Design; Process Optimization
Themenfelder/Aktivitätsfelder der BAM:Chemie und Prozesstechnik
Chemie und Prozesstechnik / Anlagensicherheit und Prozesssimulation
Veranstaltung:35th European Symposium on Computer Aided Process Engineering (ESCAPE)
Veranstaltungsort:Gent, Belgium
Beginndatum der Veranstaltung:06.07.2025
Verfügbarkeit des Dokuments:Datei im Netzwerk der BAM verfügbar ("Closed Access")
Datum der Freischaltung:17.12.2025
Referierte Publikation:Nein
Einverstanden
Diese Webseite verwendet technisch erforderliche Session-Cookies. Durch die weitere Nutzung der Webseite stimmen Sie diesem zu. Unsere Datenschutzerklärung finden Sie hier.