In view of the need to decarbonize the actual energy sector, already the degree of penetration of renewable energy sources in the existing power system has considerably increased. Nevertheless, in order to reduce the curtailment of renewables and still provide the ancillary services that ensure the efficient and safe operation of the grid, it is also necessary to include energy storage solutions in the equation. Within BES technologies, in the last years, the redox flow battery (RFB) has become a serious candidate to exploit the rapidly increasing niche market in large energy storage stationary applications due to its intrinsic characteristics: independent energy and power ratings, high round trip efficiency (RTE), depth of discharge, fast response, reduced environmental impact and a very competitive levelized cost of storage (LCOS). As occurs with other technologies, the target of reaching the industrial maturity of RFBs involves extensive and intensive experimentation work so to overcome the techno-economic barriers. In this aspect, the utilisation of different modelling and simulation techniques provide practical support in the R&D activities at various levels, from material and component design up to system operation. In this work in particular, it is sought to develop a zero-dimensional dynamic electrochemical cell model of a RFB with low computational costs in MATLAB-Simulink, with the objective of creating a tool that will serve as a reference to investigate battery operation and management strategies in the future. Starting from the study of the state of art referred to RFB modelling and simulation, based on the most common literature approaches, initially a lumped mathematical model of a RFB cell has been described, considering the concentration discrepancy of the active species in the tanks and cell, as well as, the various overpotentials (ohmic, activation and concentration) that arise from the limitations of the internal physicochemical processes. This model, referred as original model throughout the work, is experimentally validated with a single cell laboratory set-up symmetrically configured so to obtain an insight of its accuracy. Owing to the poor performance, two empirical model improvements are suggested. On one hand, an alternative approach to obtain a more realistic cell ohmic resistance by means of experimental impedance measurements is presented. On the other hand, a coupled calculation of the activation and concentration overpotentials by iterative means is also presented. The validation of the upgraded model has proven to achieve a higher level of accuracy compared to the original, giving an inkling of the potential of the experimentally hybridized models. Besides, it has also served to generate new ideas and lines of future work, as set out in the conclusions of this work.
In considerazione della necessità di decarbonizzare l'attuale settore energetico, il grado di penetrazione delle fonti energetiche rinnovabili nel sistema energetico esistente è già notevolmente aumentato. Tuttavia, al fine di ridurre la intermittenza delle energie rinnovabili e fornire comunque i servizi ausiliari che garantiscono il funzionamento efficiente e sicuro della rete ellettrica, è necessario includere nell'equazione anche soluzioni di stoccaggio dell'energia. All'interno delle tecnologie di stoccaggio, negli ultimi anni, la batteria a flusso redox è diventata un serio candidato per sfruttare la nicchia di mercato in rapida crescita nelle grandi applicazioni stazionarie grazie alle sue caratteristiche intrinseche: energia e potenza indipendenti, alta efficienza di andata e ritorno, profondità di scarico, una risposta rapida, un impatto ambientale ridotto e una competitiva LCOS. Come avviene per altre tecnologie, l'obiettivo di raggiungere la maturità industriale delle RFB prevede un lavoro di sperimentazione ampio e intenso per superare le barriere tecnico-economiche. In questo aspetto, l'utilizzo di diverse tecniche di modellazione e simulazione fornisce supporto pratico nelle attività di ricerca e sviluppo a vari livelli, dalla progettazione di materiali e componenti fino al funzionamento del sistema. In questo lavoro in particolare, si cerca di sviluppare un modello di cella elettrochimica dinamica zero dimensionale di un batteria a flusso redox con bassi costi computazionali in MATLAB-Simulink, con l'obiettivo di creare uno strumento che servirà come riferimento per studiare il funzionamento e la gestione della batteria strategie in futuro. Partendo dallo studio dello stato dell'arte riferito alla modellazione e simulazione di RFB, basato sugli approcci più comuni della letteratura, è stato inizialmente descritto un modello matematico di una cella di RFB, considerando la discrepanza di concentrazione delle specie attive nelle vasche e nella cella , così come i vari sovrapotenziali (ohmici, di attivazione e di concentrazione) che derivano dalle limitazioni dei processi fisico-chimici interni. Questo modello, indicato come modello originale in tutto il lavoro, è stato validato sperimentalmente con un sistema di una sola cella configurata simmetricamente in modo da ottenere un'idea della sua accuratezza. A causa delle scarse prestazioni, vengono suggeriti due miglioramenti del modello originale usando metodi empirici. Da un lato viene presentato un approccio alternativo per ottenere una resistenza ohmica della cella più realistica mediante misure sperimentali di impedenza. In secondo luogo, viene inoltre presentato un calcolo accoppiato dei sovrapotenziali di attivazione e concentrazione per via iterativa. La validazione finale del modello aggiornato ha dimostrato di raggiungere un livello di accuratezza superiore rispetto all'originale, dando un'idea delle potenzialità dei modelli ibridati. Inoltre, è servito anche a generare nuove idee e linee di lavoro futuro, come indicato nelle conclusioni.
Optimisation of a zero-dimensional dynamic electrochemical cell model of a redox flow battery in MATLAB-Simulink with characterization tests
BELOKI ARRONDO, AITOR
2020/2021
Abstract
In view of the need to decarbonize the actual energy sector, already the degree of penetration of renewable energy sources in the existing power system has considerably increased. Nevertheless, in order to reduce the curtailment of renewables and still provide the ancillary services that ensure the efficient and safe operation of the grid, it is also necessary to include energy storage solutions in the equation. Within BES technologies, in the last years, the redox flow battery (RFB) has become a serious candidate to exploit the rapidly increasing niche market in large energy storage stationary applications due to its intrinsic characteristics: independent energy and power ratings, high round trip efficiency (RTE), depth of discharge, fast response, reduced environmental impact and a very competitive levelized cost of storage (LCOS). As occurs with other technologies, the target of reaching the industrial maturity of RFBs involves extensive and intensive experimentation work so to overcome the techno-economic barriers. In this aspect, the utilisation of different modelling and simulation techniques provide practical support in the R&D activities at various levels, from material and component design up to system operation. In this work in particular, it is sought to develop a zero-dimensional dynamic electrochemical cell model of a RFB with low computational costs in MATLAB-Simulink, with the objective of creating a tool that will serve as a reference to investigate battery operation and management strategies in the future. Starting from the study of the state of art referred to RFB modelling and simulation, based on the most common literature approaches, initially a lumped mathematical model of a RFB cell has been described, considering the concentration discrepancy of the active species in the tanks and cell, as well as, the various overpotentials (ohmic, activation and concentration) that arise from the limitations of the internal physicochemical processes. This model, referred as original model throughout the work, is experimentally validated with a single cell laboratory set-up symmetrically configured so to obtain an insight of its accuracy. Owing to the poor performance, two empirical model improvements are suggested. On one hand, an alternative approach to obtain a more realistic cell ohmic resistance by means of experimental impedance measurements is presented. On the other hand, a coupled calculation of the activation and concentration overpotentials by iterative means is also presented. The validation of the upgraded model has proven to achieve a higher level of accuracy compared to the original, giving an inkling of the potential of the experimentally hybridized models. Besides, it has also served to generate new ideas and lines of future work, as set out in the conclusions of this work.File | Dimensione | Formato | |
---|---|---|---|
Master Thesis A.Beloki_914980.pdf
accessibile in internet per tutti
Descrizione: Master Thesis Aitor Beloki 914980
Dimensione
2.23 MB
Formato
Adobe PDF
|
2.23 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/10589/180272