Cities play a crucial role in addressing climate change through the integration of renewable energy and improvements in energy efficiency in an increasingly urbanised world. Building operations account for nearly 27% of global energy-related CO2 emissions, while the transportation sector contributes approximately one-sixth. Urban decarbonisation is therefore a key strategy for achieving a 55% reduction in emissions by 2030 and complete decarbonisation by 2050. In this context, this study develops and evaluates optimised configuration for district multi-energy systems (D-MESs) designed to enable net-zero energy districts in urban and peri-urban environments. The study proposes a co-design framework that integrates thermal distribution and electricity balance at the district scale. It begins by establishing a standardised definition of district multi-energy systems through a systematic review of technologies and terminology. A representative district multi-energy system topology is developed based on the reviewed energy components and technology portfolios. A comparative evaluation of energy modelling tools is then conducted, considering data granularity, modelling approach, and application scale. The Forlanini mixed-use district in Milan, Italy, is adopted as the case study. A robust data collection framework ensures high modelling accuracy, with building energy demand simulated in SketchUp and the Integrated Environmental Solutions Virtual Environment (IES). For the thermal distribution network, a scenario-based evaluation is performed to compare five system configurations with different building envelope renovation, heat pump system and photovoltaic (PV) system integrations. Distribution optimisation is then applied to the configurations with superior performance of HP system. A deterministic optimisation model is employed to support the design of distributed energy systems under key uncertainties, such as renewable generation intermittency, temperature variability, and occupant behaviours. Three distribution strategies are examined: a fully decentralised air-source heat pump (ASHP) system, a hybrid decentralised groundwater-source heat pump (GWHP) system, and a centralised GWHP system, used to design the distribution network to meet the district's thermal demands (heating, cooling, and domestic hot water). For the hybrid GWHP configuration, a two-stage optimisation framework combining the Minimum Spanning Tree (MST), Genetic Algorithm (GA), and Particle Swarm Optimisation (PSO) is developed to identify optimal well locations and piping layouts, supported by Monte Carlo simulations and sensitivity analyses. The optimal configuration is further assessed in terms of the electricity dispatch, identifying suitable penetration levels of PV, heat pumps, and electric vehicles (EVs). A scheduled EV charging scheme is developed to enhance the temporal alignment between PV generation and EV charging while mitigating grid stress. System performance is evaluated through a three-dimensional penetration analysis using five indicators: electricity saving ratio, self-consumption ratio, self-sufficiency ratio, and their harmonic mean. Results indicate that the MST-based optimisation produced an efficient thermal network connecting 199 buildings with a total pipe length of 9.37 km, while the PSO algorithm achieved optimal well separation aligned with regional groundwater flow. The proposed scheduled PV-EV charging scheme reduces peak electricity demand by 10-40%, and the integration analysis shows that balanced technology deployment, yields the best overall performance. The optimal configuration comprises 100% PV capacity, 40% EV adoption, and 70% GWHP penetration, achieving substantial reductions in emissions (81%) and operational costs. In summary, the results demonstrate that the interaction between energy sources and vectors occurs at multiple levels and can generate synergistic effects. Optimal system performance is achieved not by maximising individual technology penetration, but through a balanced and coordinated deployment strategy. The primary contribution of this work lies in the development of a collaborative optimisation methodology for the distribution design of district-scale multi-energy systems. The findings provide valuable insights for urban energy planners and policymakers aiming to enhance renewable integration and support the transition towards more resilient, sustainable urban energy systems.
Le città svolgono un ruolo cruciale nel contrasto al cambiamento climatico attraverso l’integrazione delle energie rinnovabili e il miglioramento dell’efficienza energetica in un mondo sempre più urbanizzato. Il funzionamento degli edifici è responsabile di circa il 27% delle emissioni globali di CO₂ legate all’energia, mentre il settore dei trasporti contribuisce per circa un sesto. La decarbonizzazione urbana rappresenta quindi una strategia chiave per conseguire una riduzione del 55% delle emissioni entro il 2030 e la completa decarbonizzazione entro il 2050. In questo contesto, il presente studio sviluppa e valuta configurazioni ottimizzate di sistemi multi-energetici di distretto (District Multi-Energy Systems, D-MES) finalizzati a consentire distretti a energia netta zero in ambiti urbani e periurbani. Lo studio propone un quadro di co-progettazione che integra, alla scala di distretto, la distribuzione termica e il bilancio elettrico. In primo luogo, viene definita una terminologia standardizzata dei D-MES mediante una revisione sistematica delle tecnologie e dei concetti. Sulla base dei componenti energetici e dei portafogli tecnologici analizzati, viene sviluppata una topologia rappresentativa di sistema multi-energetico di distretto. Successivamente, viene condotta una valutazione comparativa degli strumenti di modellazione energetica considerando la granularità dei dati, l’approccio di modellazione e la scala applicativa. Come caso studio è adottato il distretto mixed-use di Forlanini a Milano (Italia). Un solido quadro di raccolta dati garantisce un’elevata accuratezza modellistica, con la domanda energetica degli edifici simulata in SketchUp e in Integrated Environmental Solutions Virtual Environment (IES). Per la rete di distribuzione termica, viene eseguita una valutazione basata su scenari al fine di confrontare cinque configurazioni di sistema caratterizzate da diversi livelli di riqualificazione dell’involucro edilizio e da differenti integrazioni tra pompe di calore e sistemi fotovoltaici (PV). Alle configurazioni con migliori prestazioni del sistema a pompe di calore viene quindi applicata un’ottimizzazione della distribuzione. Un modello di ottimizzazione deterministico è impiegato per supportare la progettazione di sistemi energetici distribuiti in presenza di incertezze chiave, quali l’intermittenza della generazione rinnovabile, la variabilità della temperatura e i comportamenti degli occupanti. Vengono esaminate tre strategie di distribuzione: (i) un sistema completamente decentralizzato con pompe di calore aria-acqua (ASHP), (ii) un sistema ibrido decentralizzato con pompe di calore a sorgente di falda (GWHP) e (iii) un sistema centralizzato GWHP, utilizzati per progettare la rete di distribuzione in grado di soddisfare le esigenze termiche del distretto (riscaldamento, raffrescamento e acqua calda sanitaria). Per la configurazione ibrida GWHP viene sviluppato un quadro di ottimizzazione a due stadi che combina Minimum Spanning Tree (MST), Algoritmo Genetico (GA) e Particle Swarm Optimisation (PSO) per identificare le posizioni ottimali dei pozzi e i layout delle tubazioni, supportato da simulazioni Monte Carlo e analisi di sensibilità. La configurazione ottimale è ulteriormente valutata in termini di dispacciamento elettrico, individuando livelli adeguati di penetrazione di PV, pompe di calore e veicoli elettrici (EV). Viene sviluppato uno schema di ricarica programmata degli EV per migliorare l’allineamento temporale tra generazione PV e ricarica, mitigando al contempo lo stress sulla rete elettrica. Le prestazioni del sistema sono valutate tramite un’analisi tridimensionale di penetrazione tecnologica basata su cinque indicatori: rapporto di risparmio di elettricità, autoconsumo, autosufficienza e la loro media armonica. I risultati indicano che l’ottimizzazione basata su MST ha prodotto una rete termica efficiente che connette 199 edifici con una lunghezza totale delle tubazioni pari a 9,37 km, mentre l’algoritmo PSO ha ottenuto una separazione ottimale dei pozzi coerente con il flusso regionale della falda. Lo schema proposto di ricarica programmata PV–EV riduce la domanda elettrica di picco del 10–40% e l’analisi di integrazione evidenzia che una distribuzione bilanciata delle tecnologie garantisce le migliori prestazioni complessive. La configurazione ottimale prevede il 100% di capacità PV, il 40% di adozione EV e il 70% di penetrazione GWHP, conseguendo riduzioni significative delle emissioni (81%) e dei costi operativi. In sintesi, i risultati mostrano che l’interazione tra fonti e vettori energetici avviene a più livelli e può generare effetti sinergici. Le prestazioni ottimali del sistema non si ottengono massimizzando la penetrazione di ciascuna tecnologia singolarmente, bensì attraverso una strategia di implementazione bilanciata e coordinata. Il principale contributo del presente lavoro risiede nello sviluppo di una metodologia di ottimizzazione collaborativa per la progettazione della distribuzione nei sistemi multi-energetici alla scala di distretto. I risultati forniscono indicazioni utili a pianificatori energetici urbani e decisori politici per migliorare l’integrazione delle rinnovabili e supportare la transizione verso sistemi energetici urbani più resilienti e sostenibili.
Optimised configurations of multi-energy systems with high renewable energy penetration at a district scale
Xu, Yingqing
2025/2026
Abstract
Cities play a crucial role in addressing climate change through the integration of renewable energy and improvements in energy efficiency in an increasingly urbanised world. Building operations account for nearly 27% of global energy-related CO2 emissions, while the transportation sector contributes approximately one-sixth. Urban decarbonisation is therefore a key strategy for achieving a 55% reduction in emissions by 2030 and complete decarbonisation by 2050. In this context, this study develops and evaluates optimised configuration for district multi-energy systems (D-MESs) designed to enable net-zero energy districts in urban and peri-urban environments. The study proposes a co-design framework that integrates thermal distribution and electricity balance at the district scale. It begins by establishing a standardised definition of district multi-energy systems through a systematic review of technologies and terminology. A representative district multi-energy system topology is developed based on the reviewed energy components and technology portfolios. A comparative evaluation of energy modelling tools is then conducted, considering data granularity, modelling approach, and application scale. The Forlanini mixed-use district in Milan, Italy, is adopted as the case study. A robust data collection framework ensures high modelling accuracy, with building energy demand simulated in SketchUp and the Integrated Environmental Solutions Virtual Environment (IES). For the thermal distribution network, a scenario-based evaluation is performed to compare five system configurations with different building envelope renovation, heat pump system and photovoltaic (PV) system integrations. Distribution optimisation is then applied to the configurations with superior performance of HP system. A deterministic optimisation model is employed to support the design of distributed energy systems under key uncertainties, such as renewable generation intermittency, temperature variability, and occupant behaviours. Three distribution strategies are examined: a fully decentralised air-source heat pump (ASHP) system, a hybrid decentralised groundwater-source heat pump (GWHP) system, and a centralised GWHP system, used to design the distribution network to meet the district's thermal demands (heating, cooling, and domestic hot water). For the hybrid GWHP configuration, a two-stage optimisation framework combining the Minimum Spanning Tree (MST), Genetic Algorithm (GA), and Particle Swarm Optimisation (PSO) is developed to identify optimal well locations and piping layouts, supported by Monte Carlo simulations and sensitivity analyses. The optimal configuration is further assessed in terms of the electricity dispatch, identifying suitable penetration levels of PV, heat pumps, and electric vehicles (EVs). A scheduled EV charging scheme is developed to enhance the temporal alignment between PV generation and EV charging while mitigating grid stress. System performance is evaluated through a three-dimensional penetration analysis using five indicators: electricity saving ratio, self-consumption ratio, self-sufficiency ratio, and their harmonic mean. Results indicate that the MST-based optimisation produced an efficient thermal network connecting 199 buildings with a total pipe length of 9.37 km, while the PSO algorithm achieved optimal well separation aligned with regional groundwater flow. The proposed scheduled PV-EV charging scheme reduces peak electricity demand by 10-40%, and the integration analysis shows that balanced technology deployment, yields the best overall performance. The optimal configuration comprises 100% PV capacity, 40% EV adoption, and 70% GWHP penetration, achieving substantial reductions in emissions (81%) and operational costs. In summary, the results demonstrate that the interaction between energy sources and vectors occurs at multiple levels and can generate synergistic effects. Optimal system performance is achieved not by maximising individual technology penetration, but through a balanced and coordinated deployment strategy. The primary contribution of this work lies in the development of a collaborative optimisation methodology for the distribution design of district-scale multi-energy systems. The findings provide valuable insights for urban energy planners and policymakers aiming to enhance renewable integration and support the transition towards more resilient, sustainable urban energy systems.| File | Dimensione | Formato | |
|---|---|---|---|
|
Thesis_Manuscript.pdf
solo utenti autorizzati a partire dal 26/12/2026
Descrizione: Thesis Manuscript
Dimensione
13.39 MB
Formato
Adobe PDF
|
13.39 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/10589/248417