This thesis explores the feasibility of applying an electronic nose (e-nose), based on metal oxide semiconductor (MOX) gas sensors, for the detection of key volatile organic compounds (VOCs) released during the frying process. After identifying critical odor-active markers from literature, three representative compounds, 2-pentylfuran as a furan, nonanal as an aldehyde, and dimethyl trisulfide as a sulfur-containing compound, were selected to simulate frying emissions. 19 Nineteen sensors, including commercial TGS models and MOX variants, were tested using two commercial SACMI e-nose devices (EOS57 and EOS02) from Sacmi (Imola, Italy) and an additional custom-made chamber, EOS57 – Small chamber. Each sensor was exposed to 3controlled concentrations in triplicate of the selected VOCs, and the resulting resistance signals were recorded and processed using a Python-based analysis pipeline. This pipeline included baseline correction, normalization, smoothing , and feature extraction used to calculate key performance indicators characterization parameters such as sensitivity, limit of detection (LoD), repeatability, and response time (t₉₀). The experimental setup further allowed a comparative assessment of different sensor array configurations. and exposure chambers Overall, the study demonstrated the feasibility of combining different MOS sensors for VOC detection in frying environments. The obtained results are promising and provide a solid foundation for future developments in process monitoring, sensor integration, and data-driven modelling forin food engineering applications.
Questa tesi esplora la fattibilità dell'applicazione di un naso elettronico (e-nose), basato su sensori di gas a semiconduttore di ossido metallico (MOX), per la rilevazione dei principali composti organici volatili (VOC) rilasciati durante il processo di frittura. Dopo aver identificato i marcatori critici di odore attivi dalla letteratura, sono stati selezionati tre composti rappresentativi, il 2-pentilfurano come furano, il nonanale come aldeide e il trisolfuro di dimetile come composto contenente zolfo, per simulare le emissioni di frittura. 19 Diciannove sensori, tra cui modelli commerciali TGS e varianti MOX, sono stati testati utilizzando due dispositivi commerciali SACMI e-nose (EOS57 e EOS02) di Sacmi (Imola, Italia) e un'ulteriore camera personalizzata, EOS57 - Small chamber. Ogni sensore è stato esposto in triplo a concentrazioni controllate dei COV selezionati e i segnali di resistenza risultanti sono stati registrati ed elaborati con una pipeline di analisi basata su Python. Questa pipeline comprendeva la correzione della linea di base, la normalizzazione, lo smoothing e l'estrazione delle caratteristiche utilizzate per calcolare i parametri di caratterizzazione degli indicatori di prestazione chiave come la sensibilità, il limite di rilevamento (LoD), la ripetibilità e il tempo di risposta (t₉₀). L'impostazione sperimentale ha inoltre consentito una valutazione comparativa di diverse configurazioni di array di sensori. e camere di esposizione Nel complesso, lo studio ha dimostrato la fattibilità della combinazione di diversi sensori MOS per il rilevamento di VOC in ambienti di frittura. I risultati ottenuti sono promettenti e forniscono una solida base per gli sviluppi futuri nel campo della tecnologia.
E-nose for the real-time monitoring of frying: a methodological approach for studying feasibility with application-specific calibrants
KAKROODI, MOUSA
2024/2025
Abstract
This thesis explores the feasibility of applying an electronic nose (e-nose), based on metal oxide semiconductor (MOX) gas sensors, for the detection of key volatile organic compounds (VOCs) released during the frying process. After identifying critical odor-active markers from literature, three representative compounds, 2-pentylfuran as a furan, nonanal as an aldehyde, and dimethyl trisulfide as a sulfur-containing compound, were selected to simulate frying emissions. 19 Nineteen sensors, including commercial TGS models and MOX variants, were tested using two commercial SACMI e-nose devices (EOS57 and EOS02) from Sacmi (Imola, Italy) and an additional custom-made chamber, EOS57 – Small chamber. Each sensor was exposed to 3controlled concentrations in triplicate of the selected VOCs, and the resulting resistance signals were recorded and processed using a Python-based analysis pipeline. This pipeline included baseline correction, normalization, smoothing , and feature extraction used to calculate key performance indicators characterization parameters such as sensitivity, limit of detection (LoD), repeatability, and response time (t₉₀). The experimental setup further allowed a comparative assessment of different sensor array configurations. and exposure chambers Overall, the study demonstrated the feasibility of combining different MOS sensors for VOC detection in frying environments. The obtained results are promising and provide a solid foundation for future developments in process monitoring, sensor integration, and data-driven modelling forin food engineering applications.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_07_Kakroodi_Executive_Summary.pdf
non accessibile
Descrizione: Executive summary
Dimensione
1.05 MB
Formato
Adobe PDF
|
1.05 MB | Adobe PDF | Visualizza/Apri |
|
2025_07_Kakroodi_Thesis.pdf
non accessibile
Dimensione
15.22 MB
Formato
Adobe PDF
|
15.22 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/10589/240779