In recent years, the limitations of scalings in power-law form that are obtained from traditional log regression have become increasingly evident in many fields of research. Given the wide gap in operational space between present-day and next-generation devices, robustness of the obtained models in guaranteeing reasonable extrapolability is a major issue. In this paper, a new technique, called symbolic regression, is reviewed, refined, and applied to the ITPA database for extracting scaling laws of the energy-confinement time at different radiated fraction levels. The main advantage of this new methodology is its ability to determine the most appropriate mathematical form of the scaling laws to model the available databases without the restriction of their having to be power laws. In a completely new development, this technique is combined with the concept of geodesic distance on Gaussian manifolds so as to take into account the error bars in the measurements and provide more reliable models. Robust scaling laws, including radiated fractions as regressor, have been found; they are not in power-law form, and are significantly better than the traditional scalings. These scaling laws, including radiated fractions, extrapolate quite differently to ITER, and therefore they require serious consideration. On the other hand, given the limitations of the existing databases, dedicated experimental investigations will have to be carried out to fully understand the impact of radiated fractions on the confinement in metallic machines and in the next generation of devices.

Murari, A., Peluso, E., Gaudio, P., Gelfusa, M. (2017). Robust scaling laws for energy confinement time, including radiated fraction, in Tokamaks. NUCLEAR FUSION, 57(12), 126017 [10.1088/1741-4326/aa7bb4].

Robust scaling laws for energy confinement time, including radiated fraction, in Tokamaks

Peluso E.
;
Gaudio P.
Funding Acquisition
;
Gelfusa M.
Supervision
2017-01-01

Abstract

In recent years, the limitations of scalings in power-law form that are obtained from traditional log regression have become increasingly evident in many fields of research. Given the wide gap in operational space between present-day and next-generation devices, robustness of the obtained models in guaranteeing reasonable extrapolability is a major issue. In this paper, a new technique, called symbolic regression, is reviewed, refined, and applied to the ITPA database for extracting scaling laws of the energy-confinement time at different radiated fraction levels. The main advantage of this new methodology is its ability to determine the most appropriate mathematical form of the scaling laws to model the available databases without the restriction of their having to be power laws. In a completely new development, this technique is combined with the concept of geodesic distance on Gaussian manifolds so as to take into account the error bars in the measurements and provide more reliable models. Robust scaling laws, including radiated fractions as regressor, have been found; they are not in power-law form, and are significantly better than the traditional scalings. These scaling laws, including radiated fractions, extrapolate quite differently to ITER, and therefore they require serious consideration. On the other hand, given the limitations of the existing databases, dedicated experimental investigations will have to be carried out to fully understand the impact of radiated fractions on the confinement in metallic machines and in the next generation of devices.
2017
Pubblicato
Rilevanza internazionale
Articolo
Esperti anonimi
Settore FIS/01 - FISICA SPERIMENTALE
English
Con Impact Factor ISI
genetic programming, symbolic regression, geodesic distance, scaling laws, energy confinement time, radiated fraction, robust scalings
Murari, A., Peluso, E., Gaudio, P., Gelfusa, M. (2017). Robust scaling laws for energy confinement time, including radiated fraction, in Tokamaks. NUCLEAR FUSION, 57(12), 126017 [10.1088/1741-4326/aa7bb4].
Murari, A; Peluso, E; Gaudio, P; Gelfusa, M
Articolo su rivista
File in questo prodotto:
File Dimensione Formato  
Murari_2017_Nucl._Fusion_57_126017.pdf

solo utenti autorizzati

Licenza: Copyright dell'editore
Dimensione 1.41 MB
Formato Adobe PDF
1.41 MB Adobe PDF   Visualizza/Apri   Richiedi una copia

I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/2108/190937
Citazioni
  • ???jsp.display-item.citation.pmc??? ND
  • Scopus 12
  • ???jsp.display-item.citation.isi??? 10
social impact