The integration of renewable energy sources, given their intermittent nature, requires solutions to reduce the temporal mismatch between power demand and supply. Buildings' energy-flexible resources can help mitigate this issue. One readily available and cost-effective resource for energy flexibility in buildings is their thermal mass. By simply adjusting the thermostat setpoint, raising or lowering it within acceptable thermal comfort limits, thermal mass can store and release heat without requiring additional investments. However, this flexibility is highly dynamic and influenced by various factors. The objective of this study is to quantify the time-dependent energy flexibility of the building's thermal mass and assess the performance of the heat storage process. This quantification is conducted using the flexibility envelope concept, while the performance is assessed through the key performance indicator known as storage efficiency. The key innovation of this work lies in the use of a co-simulation framework that integrates a detailed physics-based model of the thermal zone and HVAC system with a Model Predictive Cotrol (MPC) algorithm. This approach provides a more realistic assessment of energy flexibility compared to previous studies that relied on simplified Resistance-Capacitance network models for the thermal zone and HVAC system. Flexibility is evaluated for two thermal zones, both heated by a floor heating system. The first one represents the case of an apartment in a multi-unit block, while the second one in an isolated two-story house. From the perspective of energy flexibility provided by the building’s thermal mass, the two case studies yield similar results. Specifically, under the most favourable Active Demand Response event, the maximum storage efficiency reaches 56% for the first case and 62% for the second. However, the thermal zone within the apartment achieves the highest amount of energy stored during the most efficient Active Demand Response event (i.e. 0.49 kWh/m^2, versus 0.40 kWh/m^2 for the isolated house). This flexibility, achieved through thermal mass storage, is equivalent to that provided by a 100-litre domestic hot water tank. This study demonstrates that the energy flexibility provided by the building’s thermal mass varies based on the event start time and fluctuates daily. It is influenced by several factors, including the initial state of charge of the building's thermal mass and external disturbances such as changing weather conditions and occupancy patterns. Lastly, I extended the results to a district scale, quantifying the flexibility provided by the 615 thermal zones with similar characteristics. On a cold, cloudy winter day, approximately 0.22 kWh/m^2 of energy is recovered from 0.52 kWh/m^2 stored (43% efficiency), while on a warm, sunny winter day, 0.15 kWh/m^2 is recovered from 0.37 kWh/m^2 stored (40% efficiency). As a concluding step, to facilitate real-world applications, a closed-loop optimization based solely on a reduced-order model of the thermal zone and HVAC system was tested. The optimization using the reduced-order model underestimates the duration for which the system can remain off. However, it performs reasonably well in estimating durations at high power levels, with deviations of 30 to 45 minutes for power values above 2.4 kW compared to the co-simulation. Conversely, for lower power levels between 0.8 and 1.4 kW, the reduced-order model overestimates operating durations, with deviations reaching 10–15 hours in some cases. Thus while the trend of flexibility is generally correct, for higher precision, the use of co-simulation is suggested.
L’integrazione delle fonti di energia rinnovabile, data la loro natura intermittente, richiede soluzioni per ridurre la discrepanza temporale tra domanda e produzione di energia. Le fonti di flessibilità energetica degli edifici possono contribuire ad attenuare questo problema. Una risorsa prontamente disponibile ed economicamente vantaggiosa per la flessibilità energetica negli edifici è la loro massa termica. Semplicemente regolando il setpoint del termostato, aumentandolo o diminuendolo entro limiti accettabili di comfort termico, la massa termica può accumulare e rilasciare calore senza richiedere investimenti aggiuntivi. Tuttavia, questa flessibilità è altamente dinamica e influenzata da diversi fattori. L’obiettivo di questo studio è quantificare la flessibilità energetica della massa termica dell’edificio in funzione del tempo e valutare le prestazioni del processo di accumulo termico. Questa quantificazione è effettuata utilizzando il concetto di "flexibility envelope", mentre la prestazione viene valutata attraverso l’efficienza di accumulo. L’elemento innovativo principale di questo lavoro risiede nell’utilizzo di un framework di co-simulazione che integra un modello fisico dettagliato della zona termica e del sistema HVAC con un algoritmo di controllo predittivo (Model Predictive Control - MPC). Questo approccio consente una valutazione più realistica della flessibilità energetica rispetto agli studi precedenti che si basavano su modelli semplificati per la zona termica e il sistema HVAC, rappresentati da reti Resistenza-Capacità. La flessibilità viene valutata per due zone termiche, entrambe riscaldate tramite un sistema di riscaldamento a pavimento. La prima rappresenta il caso di un appartamento in un complesso residenziale, mentre la seconda una villetta indipendente su due piani. Dal punto di vista della flessibilità energetica fornita dalla massa termica dell’edificio, i due casi studio forniscono risultati simili. In particolare, durante l’evento di Active Demand Response più vantaggioso, l’efficienza massima di accumulo raggiunge il 56% per il primo caso e il 62% per il secondo. Tuttavia, la zona termica dell’appartamento raggiunge la quantità più elevata di energia accumulata durante l’evento di Active Demand Response più efficiente (ossia 0.49 kWh/m^2 contro 0.40 kWh/m^2 per la villetta indipendente). Questa flessibilità, ottenuta tramite accumulo nella massa termica, è equivalente a quella fornita da un serbatoio di acqua calda sanitaria da 100 litri. Questo studio dimostra che la flessibilità energetica fornita dalla massa termica dell’edificio varia in funzione dell’orario di inizio dell’evento e presenta fluttuazioni giornaliere. Essa è influenzata da diversi fattori, tra cui lo stato iniziale di carica della massa termica dell’edificio e fattori esterni come le condizioni meteorologiche e i profili occupazionali. Infine, ho esteso i risultati su scala distrettuale, quantificando la flessibilità fornita da 615 zone termiche con caratteristiche simili. In una giornata invernale fredda e nuvolosa, si recuperano circa 0.22 kWh/m^2 di energia su 0.52 kWh/m^2 accumulati (efficienza pari al 43%), mentre in una giornata invernale tiepida e soleggiata si recuperano 0.15 kWh/m^2 su 0.37 kWh/m^2 accumulati (efficienza pari al 40%). In conclusione, per facilitare le applicazioni pratiche, è stata testata un’ottimizzazione a circuito chiuso basata esclusivamente su un modello semplificato della zona termica e del sistema HVAC. L’ottimizzazione mediante il modello semplificato sottostima la durata per la quale il sistema può rimanere spento. Tuttavia, fornisce stime ragionevolmente accurate per le durate a livelli di potenza elevati, con scostamenti rispetto alla co-simulazione compresi tra 30 e 45 minuti per valori di potenza superiori a 2.4 kW . Al contrario, per livelli di potenza inferiori, tra 0.8 e 1.4 kW, il modello semplificato sovrastima le durate operative, con scostamenti che in alcuni casi raggiungono le 10–15 ore. Pertanto, sebbene l’andamento della flessibilità sia generalmente corretto, per ottenere una precisione maggiore è consigliato l’uso della co-simulazione.
A framework for the quantification of the energy flexibility of the building thermal mass using MPC
ALESCI, ROSSELLA
2024/2025
Abstract
The integration of renewable energy sources, given their intermittent nature, requires solutions to reduce the temporal mismatch between power demand and supply. Buildings' energy-flexible resources can help mitigate this issue. One readily available and cost-effective resource for energy flexibility in buildings is their thermal mass. By simply adjusting the thermostat setpoint, raising or lowering it within acceptable thermal comfort limits, thermal mass can store and release heat without requiring additional investments. However, this flexibility is highly dynamic and influenced by various factors. The objective of this study is to quantify the time-dependent energy flexibility of the building's thermal mass and assess the performance of the heat storage process. This quantification is conducted using the flexibility envelope concept, while the performance is assessed through the key performance indicator known as storage efficiency. The key innovation of this work lies in the use of a co-simulation framework that integrates a detailed physics-based model of the thermal zone and HVAC system with a Model Predictive Cotrol (MPC) algorithm. This approach provides a more realistic assessment of energy flexibility compared to previous studies that relied on simplified Resistance-Capacitance network models for the thermal zone and HVAC system. Flexibility is evaluated for two thermal zones, both heated by a floor heating system. The first one represents the case of an apartment in a multi-unit block, while the second one in an isolated two-story house. From the perspective of energy flexibility provided by the building’s thermal mass, the two case studies yield similar results. Specifically, under the most favourable Active Demand Response event, the maximum storage efficiency reaches 56% for the first case and 62% for the second. However, the thermal zone within the apartment achieves the highest amount of energy stored during the most efficient Active Demand Response event (i.e. 0.49 kWh/m^2, versus 0.40 kWh/m^2 for the isolated house). This flexibility, achieved through thermal mass storage, is equivalent to that provided by a 100-litre domestic hot water tank. This study demonstrates that the energy flexibility provided by the building’s thermal mass varies based on the event start time and fluctuates daily. It is influenced by several factors, including the initial state of charge of the building's thermal mass and external disturbances such as changing weather conditions and occupancy patterns. Lastly, I extended the results to a district scale, quantifying the flexibility provided by the 615 thermal zones with similar characteristics. On a cold, cloudy winter day, approximately 0.22 kWh/m^2 of energy is recovered from 0.52 kWh/m^2 stored (43% efficiency), while on a warm, sunny winter day, 0.15 kWh/m^2 is recovered from 0.37 kWh/m^2 stored (40% efficiency). As a concluding step, to facilitate real-world applications, a closed-loop optimization based solely on a reduced-order model of the thermal zone and HVAC system was tested. The optimization using the reduced-order model underestimates the duration for which the system can remain off. However, it performs reasonably well in estimating durations at high power levels, with deviations of 30 to 45 minutes for power values above 2.4 kW compared to the co-simulation. Conversely, for lower power levels between 0.8 and 1.4 kW, the reduced-order model overestimates operating durations, with deviations reaching 10–15 hours in some cases. Thus while the trend of flexibility is generally correct, for higher precision, the use of co-simulation is suggested.| File | Dimensione | Formato | |
|---|---|---|---|
|
20250424_PhDThesisRossellaAlesci.pdf
accessibile in internet per tutti a partire dal 22/04/2026
Descrizione: Quantification of the time-dependent energy flexibility of the building's thermal mass
Dimensione
5.47 MB
Formato
Adobe PDF
|
5.47 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/10589/237938