ASDEX-Upgrade and JT-60SA are two tokamaks operating in Germany and Japan, respectively. They both play an important role in fusion research, since the former is the largest European experiment, while JT-60SA, operated jointly by Japan and the EU, is presently the largest superconducting tokamak in the world. This research has the broader aim of preparing the background study for the International Thermonuclear Experimental Reactor (ITER), which will be the most advanced fusion device ever built, set to begin its operation in 2035. One of the main topics of research is the Error Field (EF) which is an intrinsic characteristic that all fusion reactors are subjected to. Error Fields originate from the misalignment due to the coils manufacturing, transportation, assembly that concur to the tokamak non-axisymmetry and the subsequent non-axisymmetry of the magnetic field, and from other perturbations of the magnetic field caused by the presence of feeders, ferromagnetic materials etc. Moreover, the so-called plasma response to these non-axisymmetric fields must be taken into account. The correction of the Error Field is crucial, since these irregularities in the field can lead to plasma disruptions, which can cause real damage to the tokamak apparatus. In this thesis, a modeling workflow for time-dependent EF correction is developed and applied to ASDEX-Upgrade and JT-60SA. The MARS-F code is used to model plasma response in a series of scenarios, in order to calculate and correct the dominant EF which triggers core plasma response. This will be named overlap Error Field. In particular, for ASDEX-Upgrade, the EF will be evaluated at first with a constant plasma response, introducing the time dependency only for vacuum EF sources (i.e. currents in poloidal field coils feedthroughs). The plasma response time-dependency will be evaluated using MARS- F and included in the workflow, in order to take into account also its evolution and its effect on the Error Field. Regarding JT-60SA, an existing Simulink model is used as a starting point to develop a code, similar to the ASDEX-Upgrade case, but optimizing Error Field correction with the specific configuration of actuators of this other machine (3 × 6 saddle coils). Optimization is carried out assuming sinusoidal currents with a single toroidal Fourier harmonic in the three rows of active coils. In this part of the work, time-dependent plasma response is not taken into account since the only available data regards low performance commissioning experiments, for which time-varying plasma response is not relevant.
ASDEX-Upgrade e JT-60SA sono due tokamak operativi rispettivamente in Germania e Giappone. Entrambi svolgono un ruolo importante nella ricerca sulla fusione, poiché il primo è il più grande esperimento europeo, mentre JT-60SA, gestito congiuntamente da Giappone e UE, è attualmente il più grande tokamak superconduttore al mondo. Questa ricerca ha l’obiettivo più ampio di preparare lo studio di base per l’ International Thermonuclear Experimental Reactor (ITER), che sarà il reattore per fusione nucleare più avanzato mai costruito, il cui avvio è previsto per il 2035. Uno dei principali argomenti di ricerca è l’Error Field (EF, in italiano Campo Errore), una caratteristica intrinseca a cui sono soggetti tutti i reattori a fusione. Gli Error Fields derivano dal disallineamento dovuto alla fabbricazione, al trasporto e all’assemblaggio delle bobine, che concorrono alla mancanza di assialsimmetria del tokamak e alla conseguente non-assialsimmetria del campo magnetico, nonché da altre perturbazioni del campo magnetico causate dalla presenza di alimentatori, materiali ferromagnetici, ecc. Inoltre, è necessario tenere conto della cosiddetta plasma response (risposta di plasma) a questi campi non assialsimmetrici. La correzione dell’Error Field è cruciale, poiché queste irregolarità nel campo possono portare a perturbazioni del plasma, che possono causare danni reali all’apparato del tokamak. In questa tesi, un workflow per la modellizzazione della correzione dell’Error Field dipendente dal tempo è stato sviluppato e applicato ad ASDEX-Upgrade e JT-60SA. Il codice MARS-F viene utilizzato per modellare la risposta di plasma in una serie di scenari, al fine di calcolare e correggere l’Error Field dominante che innesca la risposta di plasma del core. Questo verrà denominato overlap Error Field. In particolare, per ASDEX-Upgrade, l’EF verrà valutato inizialmente con una risposta di plasma costante, introducendo la dipendenza temporale solo per le sorgenti di Error Field nel vuoto (ovvero le correnti necessarie alle bobine per il campo magnetico poloidale). La dipendenza temporale della risposta di plasma verrà valutata utilizzando MARS-F e inclusa nel workflow, al fine di tenere conto anche della sua evoluzione e del suo effetto sull’Error Field. Per quanto riguarda JT-60SA, un modello Simulink esistente viene utilizzato come punto di partenza per sviluppare un codice, simile al caso ASDEX-Upgrade, ma ottimizzando la correzione dell’Error Field con la specifica configurazione degli attuatori di quest’altra macchina (3 × 6 bobine a sella). L’ottimizzazione viene effettuata assumendo correnti sinusoidali con una singola armonica di Fourier toroidale nelle tre file di bobine attive. In questa parte del lavoro, la risposta temporale di plasma non viene presa in considerazione poiché gli unici dati disponibili riguardano esperimenti di messa in servizio a basse prestazioni, per i quali una risposta di plasma variabile nel tempo non è rilevante.
Optimization of model-based error field correction in tokamaks
LUGATO, ILARIA
2025/2026
Abstract
ASDEX-Upgrade and JT-60SA are two tokamaks operating in Germany and Japan, respectively. They both play an important role in fusion research, since the former is the largest European experiment, while JT-60SA, operated jointly by Japan and the EU, is presently the largest superconducting tokamak in the world. This research has the broader aim of preparing the background study for the International Thermonuclear Experimental Reactor (ITER), which will be the most advanced fusion device ever built, set to begin its operation in 2035. One of the main topics of research is the Error Field (EF) which is an intrinsic characteristic that all fusion reactors are subjected to. Error Fields originate from the misalignment due to the coils manufacturing, transportation, assembly that concur to the tokamak non-axisymmetry and the subsequent non-axisymmetry of the magnetic field, and from other perturbations of the magnetic field caused by the presence of feeders, ferromagnetic materials etc. Moreover, the so-called plasma response to these non-axisymmetric fields must be taken into account. The correction of the Error Field is crucial, since these irregularities in the field can lead to plasma disruptions, which can cause real damage to the tokamak apparatus. In this thesis, a modeling workflow for time-dependent EF correction is developed and applied to ASDEX-Upgrade and JT-60SA. The MARS-F code is used to model plasma response in a series of scenarios, in order to calculate and correct the dominant EF which triggers core plasma response. This will be named overlap Error Field. In particular, for ASDEX-Upgrade, the EF will be evaluated at first with a constant plasma response, introducing the time dependency only for vacuum EF sources (i.e. currents in poloidal field coils feedthroughs). The plasma response time-dependency will be evaluated using MARS- F and included in the workflow, in order to take into account also its evolution and its effect on the Error Field. Regarding JT-60SA, an existing Simulink model is used as a starting point to develop a code, similar to the ASDEX-Upgrade case, but optimizing Error Field correction with the specific configuration of actuators of this other machine (3 × 6 saddle coils). Optimization is carried out assuming sinusoidal currents with a single toroidal Fourier harmonic in the three rows of active coils. In this part of the work, time-dependent plasma response is not taken into account since the only available data regards low performance commissioning experiments, for which time-varying plasma response is not relevant.| File | Dimensione | Formato | |
|---|---|---|---|
|
Lugato_Ilaria.pdf
accesso aperto
Dimensione
18.48 MB
Formato
Adobe PDF
|
18.48 MB | Adobe PDF | Visualizza/Apri |
The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License
https://hdl.handle.net/20.500.12608/108195