Eurostat reports that in 2023, household appliances accounted for 14.5% of the overall energy consumption in EU households. The user's choice impacts before, during and after the usage of a household appliance. The challenge is the development of solutions that improve the efficiency of low- to medium-cost household appliances. This thesis designs and implements the Systems As Support for Simpler Systems (SysAS for SiSys) method, which integrates Virtual Sensors and Data Fusion to leverage data from High-Complexity Systems to enhance Low-Complexity ones. The experimental implementation of this method is performed on a household refrigerator with 3 case studies with increasing complexity, presenting potential energy improvements. The case studies model (i) the internal stratification, (ii) the ice formation on the evaporator and (iii) the presence of a load in a household refrigerator without additional hardware. The method has also been theoretically applied to other appliances and systems at different complexities.
Eurostat riporta che nel 2023 gli elettrodomestici hanno rappresentato il 14,5% del consumo energetico complessivo delle famiglie nell’UE. Le scelte dell’utente hanno un impatto prima, durante e dopo l’utilizzo di un elettrodomestico. La sfida consiste nello sviluppo di soluzioni che migliorino l’efficienza degli elettrodomestici a basso e medio costo. Questa tesi progetta e implementa il metodo Systems As Support for Simpler Systems (SysAS for SiSys), che integra Sensori Virtuali e Data Fusion per sfruttare i dati provenienti da sistemi ad alta complessità al fine di migliorare quelli a bassa complessità. L’implementazione sperimentale di questo metodo viene eseguita su un frigorifero domestico con tre casi di studio a complessità crescente, mostrando potenziali miglioramenti energetici. I casi di studio modellano (i) la stratificazione interna, (ii) la formazione di ghiaccio sull’evaporatore e (iii) la presenza di un carico in un frigorifero domestico senza hardware aggiuntivo. Il metodo è stato inoltre applicato teoricamente ad altri elettrodomestici e sistemi con diversi livelli di complessità.
Leveraging machine learning and data fusion to enhance energy efficiency of low-complexity systems: application on household appliances
ILARE, DENNIS
2025/2026
Abstract
Eurostat reports that in 2023, household appliances accounted for 14.5% of the overall energy consumption in EU households. The user's choice impacts before, during and after the usage of a household appliance. The challenge is the development of solutions that improve the efficiency of low- to medium-cost household appliances. This thesis designs and implements the Systems As Support for Simpler Systems (SysAS for SiSys) method, which integrates Virtual Sensors and Data Fusion to leverage data from High-Complexity Systems to enhance Low-Complexity ones. The experimental implementation of this method is performed on a household refrigerator with 3 case studies with increasing complexity, presenting potential energy improvements. The case studies model (i) the internal stratification, (ii) the ice formation on the evaporator and (iii) the presence of a load in a household refrigerator without additional hardware. The method has also been theoretically applied to other appliances and systems at different complexities.| File | Dimensione | Formato | |
|---|---|---|---|
|
Ilare_Leveraging_machine_learning_and_Data_Fusion_to_enhance_energy_efficiency_of_Low_Complexity_systems_application_on_household_appliances.pdf
accessibile in internet per tutti
Dimensione
6.39 MB
Formato
Adobe PDF
|
6.39 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/10589/255417