Currently, data centers play a crucial role in the IT sector and their expansion is growing significantly. In 2020, their power consumption accounted for 1% of global consumption, with a forecast to reach 4% by 2030. The European Union has therefore promoted various initiatives, including the "Shaping Europe's Digital Future" programme, which aims to make data center climate neutral by 2030. The purpose of this thesis is to verify the effectiveness of some retrofitting operations on a tertiary building and to analyse the energy and economic feasibility of a system capable of recovering waste heat from a server farm to heat building’s areas. Starting from an energy audit, an energy model able to estimate building consumption on an hourly basis was obtained. In order to calibrate the model, a study of the building was conducted, consulting also external environmental conditions and historical consumption. By applying different boundary conditions, the resilience of the building to climate change was tested. The results obtained show that against the calibrated model, the other scenarios show higher consumption, with increases of up to 30% per year. The building retrofit, which consists of the replacement of some transparent and opaque structures of the envelope, providing greater thermal insulation, was also tested. Simulations show reductions between 10% and 15%. The second part of the thesis concerns the estimation of the waste heat recovery from a real server farm case study. After conducting an analysis on the operation of the data center and estimating the waste heat, the sizing of the recovery system and the reused heat have been calculated. The estimation was carried out for both the calibrated model and the two scenarios. The procedure was also reproduced for a hypothetical expansion of the data center, increasing the installed server power by 50% and consequently, the waste heat. The analysis yielded positive energy and financial results in both cases: the reduction of boiler fuel consumption range between 65% and 74% in the current state and between 84% and 92% with the expansion. Considering a 20-year time horizon, current scenario shows an NPV between 175 and 188 k€, whereas with the extension, a value of 250-277 k€ is reached. The PBT also decreases from 6 to 5 years.
Attualmente, i data center svolgono un ruolo cruciale nel settore IT e la loro espansione è in forte crescita. Nel 2020, il loro consumo di energia elettrica ha rappresentato l'1% del consumo globale, con una previsione di raggiungere il 4% entro il 2030. L'Unione Europea ha pertanto promosso varie iniziative, tra cui il programma “Shaping Europe's Digital Future”, che mira a rendere i data center climaticamente neutrali entro il 2030. L'obbiettivo dell'elaborato è quello di verificare l'efficacia di alcuni interventi di riqualificazione su un edificio terziario e di analizzare la fattibilità energetica ed economica di un impianto in grado di recuperare il calore di scarto da una server farm per riscaldare le aree dell’edificio. Partendo da una diagnosi energetica, si è ottenuto un modello energetico in grado di stimare i consumi dell'edificio su base oraria. Per calibrare il modello è stato condotto uno studio dell'edificio, consultando inoltre sia le condizioni ambientali esterne, sia i consumi storici. Applicando differenti condizioni al contorno, si è voluto testare la resilienza dell’edificio ai cambiamenti climatici. I risultati ottenuti mostrano che a fronte del modello calibrato, gli altri scenari presentano consumi maggiori, con aumenti fino al 30% annui. È stato inoltre verificato il retrofit edilizio che consiste nella sostituzione parziale di strutture trasparenti e opache dell’involucro, garantendo un maggior isolamento termico. Le simulazioni registrano riduzioni della domanda energetica comprese tra 10% e 15%. La seconda parte dell’elaborato riguarda la stima del recupero di calore dalla server farm per il riscaldamento di un edificio terziario. Dopo aver condotto un'analisi sul funzionamento del data center e stimato il calore di scarto, si è calcolato il dimensionamento dell'impianto di recupero e il calore effettivamente riutilizzabile. La stima è stata effettuata sia per il modello calibrato sia per i due scenari alternativi. Il procedimento è stato riprodotto anche per un ipotetico ampliamento del data center, aumentando la potenza dei server installata del 50% e di conseguenza il calore di scarto. L'analisi ha fornito risultati positivi in termini energetici e finanziari in entrambi i casi: la riduzione del consumo di combustibile della caldaia varia tra il 65% e il 74% nello stato attuale e tra l'84% e il 92% con l’ampliamento. Sotto il profilo finanziario, considerando un orizzonte temporale di 20 anni, lo scenario attuale mostra un Net Present Value tra 175 e 188 k€ e un Pay Back Time di 6 anni, mentre con l’ampliamento l’NPV sale a 250-277 k€ e il PBT scende a 5 anni.
Recupero calore da un data center per il riscaldamento di un edificio del terziario
Pacileo, Luca
2023/2024
Abstract
Currently, data centers play a crucial role in the IT sector and their expansion is growing significantly. In 2020, their power consumption accounted for 1% of global consumption, with a forecast to reach 4% by 2030. The European Union has therefore promoted various initiatives, including the "Shaping Europe's Digital Future" programme, which aims to make data center climate neutral by 2030. The purpose of this thesis is to verify the effectiveness of some retrofitting operations on a tertiary building and to analyse the energy and economic feasibility of a system capable of recovering waste heat from a server farm to heat building’s areas. Starting from an energy audit, an energy model able to estimate building consumption on an hourly basis was obtained. In order to calibrate the model, a study of the building was conducted, consulting also external environmental conditions and historical consumption. By applying different boundary conditions, the resilience of the building to climate change was tested. The results obtained show that against the calibrated model, the other scenarios show higher consumption, with increases of up to 30% per year. The building retrofit, which consists of the replacement of some transparent and opaque structures of the envelope, providing greater thermal insulation, was also tested. Simulations show reductions between 10% and 15%. The second part of the thesis concerns the estimation of the waste heat recovery from a real server farm case study. After conducting an analysis on the operation of the data center and estimating the waste heat, the sizing of the recovery system and the reused heat have been calculated. The estimation was carried out for both the calibrated model and the two scenarios. The procedure was also reproduced for a hypothetical expansion of the data center, increasing the installed server power by 50% and consequently, the waste heat. The analysis yielded positive energy and financial results in both cases: the reduction of boiler fuel consumption range between 65% and 74% in the current state and between 84% and 92% with the expansion. Considering a 20-year time horizon, current scenario shows an NPV between 175 and 188 k€, whereas with the extension, a value of 250-277 k€ is reached. The PBT also decreases from 6 to 5 years.File | Dimensione | Formato | |
---|---|---|---|
2024_12_Pacileo.pdf
non accessibile
Descrizione: testo tesi
Dimensione
7.73 MB
Formato
Adobe PDF
|
7.73 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/10589/230071