Error analysis of dimensionless scaling experiments with multiple points using linear regression - Archive ouverte HAL Accéder directement au contenu
Article Dans Une Revue Nuclear Fusion Année : 2010

Error analysis of dimensionless scaling experiments with multiple points using linear regression

Özgür D. Gürcan
Laure Vermare
Pascale Hennequin
C. Bourdelle

Résumé

A general method of error estimation in the case of multiple point dimensionless scaling experiments, using linear regression and standard error propagation, is proposed. The method reduces to the previous result of Cordey (2009 Nucl. Fusion 49 052001) in the case of a two-point scan. On the other hand, if the points follow a linear trend, it explains how the estimated error decreases as more points are added to the scan. Based on the analytical expression that is derived, it is argued that for a low number of points, adding points to the ends of the scanned range, rather than the middle, results in a smaller error estimate.

Dates et versions

hal-01551005 , version 1 (29-06-2017)

Identifiants

Citer

Özgür D. Gürcan, Laure Vermare, Pascale Hennequin, C. Bourdelle. Error analysis of dimensionless scaling experiments with multiple points using linear regression. Nuclear Fusion, 2010, 50, pp.022003. ⟨10.1088/0029-5515/50/2/022003⟩. ⟨hal-01551005⟩
89 Consultations
0 Téléchargements

Altmetric

Partager

Gmail Facebook X LinkedIn More