The growing influence of Digital Twin (DT) technology and advanced simulation software is transforming traditional food and beverage industries through dynamic simulation tools, data-driven decision-making, and opportunities for process optimization. This thesis investigates the dynamic simulation of the distillation process in Cognac production using the DYNSIM® software. The main goal is to digitalize traditional technology by modeling the Charentais alambic still, focusing on accurately reproducing the second distillation stage. Using DYNSIM® tools, a detailed process flowsheet was developed, covering heating, reflux, product separation, and shutdown phases. Basic control strategies were implemented to manage key process variables. Reliable models were created for each unit, ensuring that no numerical errors occur throughout the simulation, and using the NRTL (Non-Random Two-Liquid) thermodynamic model. According to classical technology, 2,500 liters of a water-ethanol mixture representing low wine were virtually distilled. The output was separated into four fractions based on alcohol content. Simulation results were presented in graphs and tables illustrating key stages and system dynamics. DYNSIM® successfully replicated the traditional distillation process, capturing the system’s dynamic behavior and showing a strong correlation with real-world patterns, including the expected volatility sequence and compound-specific trends. A special focus was placed on the behavior of five key volatile compounds—acetaldehyde, ethyl acetate, methanol, isoamyl alcohol, and furfural—selected for their sensory impact, relevance to product quality and varying volatilities. Their distribution across fractions was analyzed, and compared with literature data, confirming the simulation’s ability to replicate complex, time-dependent operations. Due to lower costs and shorter time required, simulation proves a competitive alternative to physical experiments. As distillation remains an energy-intensive operation in the industry, the study helps to find ways for optimization, improving efficiency, sustainability, and product quality for classical technology. Finally, the thesis highlights the potential for expanding the model with additional components, control strategies, and energy-saving techniques, thereby demonstrating the feasibility of future digital transformation in the spirits industry.
L’influenza crescente della tecnologia del Digital Twin (DT) e dei software di simulazione avanzata sta trasformando le industrie tradizionali alimentari e delle bevande, grazie all’integrazione di strumenti di simulazione dinamica, decisioni basate sui dati e nuove opportunità di ottimizzazione dei processi. Questa tesi analizza la simulazione dinamica del processo di distillazione nella produzione del Cognac utilizzando il software DYNSIM®. L’obiettivo principale è digitalizzare la tecnologia tradizionale modellando l’alambicco Charentais, con particolare attenzione alla riproduzione accurata della seconda fase di distillazione. Attraverso il software DYNSIM® è stato sviluppato un diagramma di flusso dettagliato del processo, includendo le fasi di riscaldamento, riflusso, separazione del prodotto e arresto. Sono state implementate strategie di controllo di base per gestire le variabili di processo principali. Modelli affidabili sono stati creati per ogni unità e operazione fisica, evitando errori numerici durante la simulazione, grazie all’utilizzo del modello termodinamico NRTL (Non-Random Two-Liquid). Secondo la tecnologia classica, sono stati distillati in modo condizionale 2.500 litri di una miscela acqua-etanolo rappresentativa del “brouillis”. Il prodotto ottenuto è stato suddiviso in quattro frazioni distinte in base al contenuto alcolico. I risultati della simulazione sono stati presentati attraverso grafici e tabelle che illustrano le fasi principali e la dinamica del sistema. DYNSIM® ha riprodotto con successo il processo di distillazione tradizionale, catturando il comportamento dinamico del sistema e mostrando una forte correlazione con i modelli reali, inclusa la sequenza di volatilità attesa e le tendenze specifiche dei composti. Particolare attenzione è stata dedicata a cinque composti volatili chiave—acetaldeide, acetato di etile, metanolo, alcol isoamilico e furfurale—scelti per il loro impatto sensoriale, rilevanza nella qualità del prodotto e diversa volatilità. La loro distribuzione tra le frazioni è stata analizzata e confrontata con dati presenti in letteratura, confermando la capacità della simulazione di riprodurre operazioni complesse e dipendenti dal tempo. Grazie ai costi ridotti e ai tempi brevi di analisi, la simulazione si dimostra un’alternativa competitiva agli esperimenti fisici. Poiché la distillazione è ancora oggi una delle operazioni più energivore, lo studio contribuisce all’identificazione di strategie di ottimizzazione, migliorando efficienza, sostenibilità e qualità del prodotto nella tecnologia tradizionale. Infine, la tesi evidenzia il potenziale di espansione del modello attraverso l’integrazione di nuovi componenti, strategie di controllo avanzate e tecniche di risparmio energetico, dimostrando la fattibilità della trasformazione digitale nel settore degli alcolici.
Dynamic simulation of batch distillation in the Cognac production process
Khasikova, Anna
2024/2025
Abstract
The growing influence of Digital Twin (DT) technology and advanced simulation software is transforming traditional food and beverage industries through dynamic simulation tools, data-driven decision-making, and opportunities for process optimization. This thesis investigates the dynamic simulation of the distillation process in Cognac production using the DYNSIM® software. The main goal is to digitalize traditional technology by modeling the Charentais alambic still, focusing on accurately reproducing the second distillation stage. Using DYNSIM® tools, a detailed process flowsheet was developed, covering heating, reflux, product separation, and shutdown phases. Basic control strategies were implemented to manage key process variables. Reliable models were created for each unit, ensuring that no numerical errors occur throughout the simulation, and using the NRTL (Non-Random Two-Liquid) thermodynamic model. According to classical technology, 2,500 liters of a water-ethanol mixture representing low wine were virtually distilled. The output was separated into four fractions based on alcohol content. Simulation results were presented in graphs and tables illustrating key stages and system dynamics. DYNSIM® successfully replicated the traditional distillation process, capturing the system’s dynamic behavior and showing a strong correlation with real-world patterns, including the expected volatility sequence and compound-specific trends. A special focus was placed on the behavior of five key volatile compounds—acetaldehyde, ethyl acetate, methanol, isoamyl alcohol, and furfural—selected for their sensory impact, relevance to product quality and varying volatilities. Their distribution across fractions was analyzed, and compared with literature data, confirming the simulation’s ability to replicate complex, time-dependent operations. Due to lower costs and shorter time required, simulation proves a competitive alternative to physical experiments. As distillation remains an energy-intensive operation in the industry, the study helps to find ways for optimization, improving efficiency, sustainability, and product quality for classical technology. Finally, the thesis highlights the potential for expanding the model with additional components, control strategies, and energy-saving techniques, thereby demonstrating the feasibility of future digital transformation in the spirits industry.File | Dimensione | Formato | |
---|---|---|---|
Thesis Anna Khasikova .pdf
accessibile in internet per tutti
Descrizione: Text of the thesis
Dimensione
1.9 MB
Formato
Adobe PDF
|
1.9 MB | Adobe PDF | Visualizza/Apri |
Executive_Summary Anna Khasikova Thesis.pdf
accessibile in internet per tutti
Descrizione: Executive summary Anna Khasikova Thesis
Dimensione
1.29 MB
Formato
Adobe PDF
|
1.29 MB | 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/10589/239237