The purpose of this M.Sc. thesis work is to develop an optimization algorithm for the downstream separation of the minimum boiling azeotrope between dimethyl carbonate (DMC) and methanol through extractive distillation, with five different high boiling entrainers to be ranked according to their efficiency. This study is carried out in collabo- ration with the LCA ("Laboratoire de Chimie Agro-industrielle") and the LGC ("Labo- ratoire de Génie Chimique") groups of "École nationale supérieure des ingénieurs en arts chimiques et technologiques" of Toulouse, to contribute to Prof. Ivonne Rodríguez-Donis’ publication "Entrainer selection using the Infinitely Sharp Split method and thermodynamic criteria for separating binary minimum boiling azeotrope by extractive distillation". The objective function is given by the total annualized costs (TAC) while variables to be optimized are entrainer feed flow rate togheter with total number of trays, reflux ratio and inlet feed trays positions for both extractive and entrainer regeneration columns. In lieu of using a genetic algorithm (GA), the mixed-integer non-linear programming (MINLP) problem is solved running Aspen Plus® simulator built-in sequential quadratic programming (SQP) method for each set of discrete variable values, reducing considerably computational times. The results of the optimization considering methyl salicylate, phenol, methyl isobutyl ketone, ethyl benzoate and 2-ethoxyethanol as entrainers are validating and validated by a pre-ranking method thermodynamics-based. Moreover, from the narrow range of values obtained for the ratio FE/FE,min, it is possible to infer an heuristic rule similar to the one used for the calculation of R/Rmin in binary distillation columns.
Lo scopo di questa tesi è lo sviluppo di un algoritmo per l’ottimizzazione della separazione a valle dell’azeotropo di minimo tra il dimetilcarbonato (DMC) e il metanolo attraverso la distillazione estrattiva, con cinque diversi solventi altobollenti da classificare secondo la loro efficienza. Questo studio viene effettuato in collaborazione con i gruppi LCA ("Laboratoire de Chimie Agro-industrielle") e LGC ("Laboratoire de Génie Chimique") del "École nationale supérieure des ingénieurs en arts chimiques et technologiques" di Toulouse, per contribuire all’articolo della Prof. Ivonne Rodríguez-Donis "Entrainer selection using the Infinitely Sharp Split method and thermodynamic criteria for separating binary minimum boiling azeotrope by extractive distillation". La funzione obiettivo è data dai costi annualizzati totali (TAC) mentre le variabili da ottimizzare sono la portata di solvente insieme al numero di piatti totale, il rapporto di riflusso e le posizioni dei piatti di ingresso sia per la colonna di estrazione che per quella di rigenerazione del solvente. Al posto di utlizzare un algoritmo genetico (GA), il problema di programmazione non lineare e variabili integrali (MINLP) è risolto eseguendo il metodo di programmazione sequenziale quadratica (SQP) originario di Aspen Plus® per ogni insieme di variabili discrete, riducendo considerevolmente i tempi di calcolo. I risultati dell’ottimizzazione considerando il salicilato di metile, il fenolo, il metilisobutilchetone, il benzoato di etile e il 2-etossietanolo come solventi validano e sono validati da un metodo per la pre-selezione basato sulla termodinamica. In aggiunta, dall’intervallo ristretto di valori ottenuto per il rapporto FE/FE,min, è possibile dedurre una regola euristica simile a quella in uso per il calcolo di R/Rmin nelle colonne di distillazione binarie.
Development of a SQP algorithm for the optimization and comparison of extractive distillation designs with application to the DMC-methanol case study
Parascandolo, Edoardo
2022/2023
Abstract
The purpose of this M.Sc. thesis work is to develop an optimization algorithm for the downstream separation of the minimum boiling azeotrope between dimethyl carbonate (DMC) and methanol through extractive distillation, with five different high boiling entrainers to be ranked according to their efficiency. This study is carried out in collabo- ration with the LCA ("Laboratoire de Chimie Agro-industrielle") and the LGC ("Labo- ratoire de Génie Chimique") groups of "École nationale supérieure des ingénieurs en arts chimiques et technologiques" of Toulouse, to contribute to Prof. Ivonne Rodríguez-Donis’ publication "Entrainer selection using the Infinitely Sharp Split method and thermodynamic criteria for separating binary minimum boiling azeotrope by extractive distillation". The objective function is given by the total annualized costs (TAC) while variables to be optimized are entrainer feed flow rate togheter with total number of trays, reflux ratio and inlet feed trays positions for both extractive and entrainer regeneration columns. In lieu of using a genetic algorithm (GA), the mixed-integer non-linear programming (MINLP) problem is solved running Aspen Plus® simulator built-in sequential quadratic programming (SQP) method for each set of discrete variable values, reducing considerably computational times. The results of the optimization considering methyl salicylate, phenol, methyl isobutyl ketone, ethyl benzoate and 2-ethoxyethanol as entrainers are validating and validated by a pre-ranking method thermodynamics-based. Moreover, from the narrow range of values obtained for the ratio FE/FE,min, it is possible to infer an heuristic rule similar to the one used for the calculation of R/Rmin in binary distillation columns.File | Dimensione | Formato | |
---|---|---|---|
2023_12_Parascandolo_01.pdf
solo utenti autorizzati dal 25/11/2024
Descrizione: Thesis
Dimensione
11.46 MB
Formato
Adobe PDF
|
11.46 MB | Adobe PDF | Visualizza/Apri |
2023_12_Parascandolo_02.pdf
solo utenti autorizzati dal 25/11/2024
Descrizione: Executive Summary
Dimensione
908.35 kB
Formato
Adobe PDF
|
908.35 kB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/213432