TY - JOUR
T1 - Neutron spectra unfolding with minimum fisher regularisation
AU - Mlynar, J.
AU - Bertalot, L.
AU - Tsalas, M.
AU - Bonheure, G.
AU - Conroy, S.
N1 - Publisher Copyright:
© Copyright owned by the author(s) under the terms of the Creative Commons Attribution-NonCommercial-ShareAlike Licence.
PY - 2006
Y1 - 2006
N2 - During past experimental campaigns at JET, the NE213 liquid scintillator proved to be a valid diagnostic tool for spectral measurements of fusion neutrons emitted from the plasma. The scintillator response function to neutrons is derived from theoretical estimation combined with measurements at an accelerator-based neutron source. The spectrum unfolding poses an illconditioned inversion problem similar to tomographic reconstruction. Minimum Fisher Regularisation (MFR) is one of the established methods in plasma 2D tomography, providing a rapid and robust tool for sparse data inversion without special a priori assumptions. In neutron analyses at JET, the MFR has been applied both in tomography (see contribution by G. Bonheure) and in unfolding the NE213 spectra, where good agreement with results of the standard Maximum Entropy unfolding procedure (MAXED) was demonstrated. This contribution is focussed on further progress in the MFR application to the unfolding of NE213 spectra. In particular, the L-curve maximum curvature optimisation was implemented which gives MFR the full independence required for automated operation. New results are presented in comparison with the MAXED analyses. The differences between the data and the retrofit of the unfolded spectra are studied and the MFR potential for future applications is outlined.
AB - During past experimental campaigns at JET, the NE213 liquid scintillator proved to be a valid diagnostic tool for spectral measurements of fusion neutrons emitted from the plasma. The scintillator response function to neutrons is derived from theoretical estimation combined with measurements at an accelerator-based neutron source. The spectrum unfolding poses an illconditioned inversion problem similar to tomographic reconstruction. Minimum Fisher Regularisation (MFR) is one of the established methods in plasma 2D tomography, providing a rapid and robust tool for sparse data inversion without special a priori assumptions. In neutron analyses at JET, the MFR has been applied both in tomography (see contribution by G. Bonheure) and in unfolding the NE213 spectra, where good agreement with results of the standard Maximum Entropy unfolding procedure (MAXED) was demonstrated. This contribution is focussed on further progress in the MFR application to the unfolding of NE213 spectra. In particular, the L-curve maximum curvature optimisation was implemented which gives MFR the full independence required for automated operation. New results are presented in comparison with the MAXED analyses. The differences between the data and the retrofit of the unfolded spectra are studied and the MFR potential for future applications is outlined.
UR - http://www.scopus.com/inward/record.url?scp=85019531277&partnerID=8YFLogxK
M3 - Conference article
AN - SCOPUS:85019531277
SN - 1824-8039
JO - Proceedings of Science
JF - Proceedings of Science
T2 - 2006 International Workshop on Fast Neutron Detectors and Applications, FNDA 2006
Y2 - 3 April 2006 through 6 April 2006
ER -