High-penetration renewable generation increases short-term variability and cycling de- mands at the distribution level, making battery energy storage systems (BESS) a key flexibility resource in grid-connected microgrids. While BESS sizing and control are often discussed through energy and power metrics, thermal behaviour is a limiting factor for safety margin, efficiency, and lifetime under rapidly changing PV, wind, and residential load conditions. Early-stage technology screening is further complicated by inconsistent electrical characterization across chemistries, particularly when internal resistance must be inferred from impedance data rather than measured as pulse-defined DC internal re- sistance (DCIR). This thesis develops and implements a system-level electro-thermal simulation frame- work for a grid-connected BESS operating under time-varying PV, wind, and residential load profiles. The approach couples an energy management strategy and a bidirectional power interface with a transparent 0D lumped-capacitance thermal model. The net micro- grid power balance is translated into a battery current trajectory that drives irreversible Joule heating. To enable like-for-like cross-chemistry comparison at fixed pack targets, chemistry-dependent pack mass is derived from system-level energy density, and pack re- sistance is constructed from representative manufacturer-reported cell specifications. The framework is applied to five chemistries: lithium iron phosphate (LFP), nickel manganese cobalt (NMC), lithium titanate (LTO), sodium-ion, and solid-state lithium batteries. Results show that, under identical electrical loading, predicted thermal severity is gov- erned primarily by the effective DC resistance used in the I2R loss term. When only high-frequency impedance metrics are available, mapping from AC impedance to an ef- fective DC resistance is required for realistic loss estimation, yet this mapping becomes the dominant uncertainty in peak temperature and thermal margin. A moderate map- ping factor supported by timescale-based evidence reduces predicted peak temperatures relative to a conservative mapping, but remaining exceedance of the maximum operat- ing temperature Tmax,op in several cases indicates that thermal compliance can require reduced power rating, increased energy capacity, or improved thermal rejection. The sodium-ion baseline, relying on direct DCIR, provides a cleaner reference for resistance definition and indicates that realistic DCIR values may still lead to Tmax,op exceedance under aggressive power demands. For solid-state batteries, the bounded analysis confirms that firm thermal conclusions are not yet defensible without verified resistance charac- terization at relevant timescales. However, the results also indicate that if solid-state internal resistance decreases as the technology matures, solid-state BESS could become particularly advantageous in warm environments where operating temperature margins are most constrained.
L’elevata penetrazione delle fonti rinnovabili introduce variabilità su scala temporale breve e profili di carico fortemente ciclici nelle reti di distribuzione. In questo contesto, i sistemi di accumulo elettrochimico (Battery Energy Storage Systems, BESS) rappresentano una risorsa di flessibilità essenziale per microreti connesse alla rete. Oltre a energia e potenza, il comportamento termico costituisce un vincolo determinante, poiché influenza margini di sicurezza, rendimento e durata operativa quando fotovoltaico, eolico e carichi residenziali variano rapidamente. Nelle fasi preliminari di progettazione, la selezione tecnologica è resa più complessa dalla mancanza di una caratterizzazione elettrica coerente tra chimiche diverse, soprattutto quando la resistenza interna deve essere stimata a partire da grandezze di impedenza anziché misurata come resistenza DC tramite prove impulsive (DCIR). La tesi propone e implementa un framework di simulazione elettro-termica a livello di sistema per un BESS in una microrete connessa alla rete, soggetta a profili temporali vari- abili di fotovoltaico, eolico e domanda residenziale. L’approccio integra una strategia di energy management e un’interfaccia di potenza bidirezionale con un modello termico 0D a capacità concentrata, scelto per garantire trasparenza e rapidità computazionale. Il bi- lancio di potenza della microrete viene convertito in una traiettoria di corrente di batteria, dalla quale si stimano le perdite irreversibili per effetto Joule. Per consentire un confronto omogeneo a parità di obiettivi di pacco, la massa del pacco è derivata dalla densità en- ergetica a livello di sistema e la resistenza del pacco è costruita a partire da specifiche rappresentative riportate dai produttori. Il framework è applicato a cinque chimiche: litio- ferro-fosfato (LFP), nichel-manganese-cobalto (NMC), titanato di litio (LTO), sodio-ione e batterie al litio allo stato solido. I risultati indicano che, a parità di sollecitazione elettrica, la severità termica prevista dipende principalmente dalla resistenza DC effettiva impiegata nel termine di perdita I2R. Quando sono disponibili solo metriche di impedenza ad alta frequenza, è necessario ricondurle a una resistenza equivalente in DC per stimare le perdite in modo realistico; tuttavia, tale conversione introduce l’incertezza dominante nella temperatura di picco e nel margine termico. Un fattore di conversione moderato, motivato da considerazioni di scala temporale, riduce le temperature di picco rispetto a un’ipotesi più conservativa, ma in diversi casi permane il superamento della temperatura massima operativa Tmax,op, suggerendo la necessità di ridurre la potenza specifica, aumentare la capacità energetica o migliorare lo smaltimento termico. La chimica sodio-ione, basata su DCIR diretto, fornisce un riferimento più pulito per la definizione della resistenza e mostra che valori realistici di DCIR possono comunque portare al superamento di Tmax,op sotto richieste di potenza aggressive. Per le batterie allo stato solido, l’analisi per limiti evidenzia che conclusioni termiche robuste richiedono una caratterizzazione elettrica verificata alle scale temporali rilevanti. Al tempo stesso, emerge che una riduzione della resistenza interna, attesa con la maturazione della tecnologia, potrebbe rendere lo stato solido particolarmente interessante in ambienti caldi, dove il margine rispetto ai limiti operativi è più critico.
Electro thermal assessment of BESS in renewable microgrids
KOCAN, AYKUT
2025/2026
Abstract
High-penetration renewable generation increases short-term variability and cycling de- mands at the distribution level, making battery energy storage systems (BESS) a key flexibility resource in grid-connected microgrids. While BESS sizing and control are often discussed through energy and power metrics, thermal behaviour is a limiting factor for safety margin, efficiency, and lifetime under rapidly changing PV, wind, and residential load conditions. Early-stage technology screening is further complicated by inconsistent electrical characterization across chemistries, particularly when internal resistance must be inferred from impedance data rather than measured as pulse-defined DC internal re- sistance (DCIR). This thesis develops and implements a system-level electro-thermal simulation frame- work for a grid-connected BESS operating under time-varying PV, wind, and residential load profiles. The approach couples an energy management strategy and a bidirectional power interface with a transparent 0D lumped-capacitance thermal model. The net micro- grid power balance is translated into a battery current trajectory that drives irreversible Joule heating. To enable like-for-like cross-chemistry comparison at fixed pack targets, chemistry-dependent pack mass is derived from system-level energy density, and pack re- sistance is constructed from representative manufacturer-reported cell specifications. The framework is applied to five chemistries: lithium iron phosphate (LFP), nickel manganese cobalt (NMC), lithium titanate (LTO), sodium-ion, and solid-state lithium batteries. Results show that, under identical electrical loading, predicted thermal severity is gov- erned primarily by the effective DC resistance used in the I2R loss term. When only high-frequency impedance metrics are available, mapping from AC impedance to an ef- fective DC resistance is required for realistic loss estimation, yet this mapping becomes the dominant uncertainty in peak temperature and thermal margin. A moderate map- ping factor supported by timescale-based evidence reduces predicted peak temperatures relative to a conservative mapping, but remaining exceedance of the maximum operat- ing temperature Tmax,op in several cases indicates that thermal compliance can require reduced power rating, increased energy capacity, or improved thermal rejection. The sodium-ion baseline, relying on direct DCIR, provides a cleaner reference for resistance definition and indicates that realistic DCIR values may still lead to Tmax,op exceedance under aggressive power demands. For solid-state batteries, the bounded analysis confirms that firm thermal conclusions are not yet defensible without verified resistance charac- terization at relevant timescales. However, the results also indicate that if solid-state internal resistance decreases as the technology matures, solid-state BESS could become particularly advantageous in warm environments where operating temperature margins are most constrained.| File | Dimensione | Formato | |
|---|---|---|---|
|
Thesis_Aykut_Kocan.pdf
non accessibile
Descrizione: Electro Thermal Assessment of BESS in Renewable Microgrids
Dimensione
4.36 MB
Formato
Adobe PDF
|
4.36 MB | Adobe PDF | Visualizza/Apri |
|
Executive_Summary_Aykut_Kocan_V2.pdf
non accessibile
Descrizione: Execitive Summary Electro Thermal Assessment of BESS in Renewable Microgrids
Dimensione
1.24 MB
Formato
Adobe PDF
|
1.24 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/10589/252861