Sensitivity and uncertainty analyses of the HCLL mock-up experiment

• Published in: Fusion engineering and design Vol. 85; no. 10; pp. 1724 - 1727
• Main Authors: Leichtle, D., Fischer, U., Kodeli, I., Perel, R.L., Klix, A., Batistoni, P., Villari, R.
• Format: Journal Article Conference Proceeding
• Language: English
• Published: Amsterdam Elsevier B.V 01.12.2010; Elsevier
• Subjects: Applied sciences; Assessments; Blanketing; Computer programs; Confidence intervals; Controled nuclear fusion plants; Energy; Energy. Thermal use of fuels; Exact sciences and technology; Installations for energy generation and conversion: thermal and electrical energy; Libraries; Lithium; Monte Carlo methods; Neutronics; Nuclear cross-sections; Transport; Tritium breeder blanket; Tritium production rate; Uncertainty; Tritium breeder blanket; Neutronics; Tritium production rate; Nuclear cross-sections; Tokamak; Breeder; Lithium; Tritium; Helium; Cross section (collision); Nuclear fusion reactor; Blanket; Deterministic approach
• ISSN: 0920-3796; 1873-7196
• DOI: 10.1016/j.fusengdes.2010.05.023

Within the European Fusion Technology Programme dedicated computational methods, tools and data have been developed and validated for sensitivity and uncertainty analyses of fusion neutronics experiments. The present paper is devoted to this kind of analyses on the recent neutronics experiment on a mock-up of the Helium-Cooled Lithium Lead Test Blanket Module for ITER at the Frascati neutron generator. They comprise both probabilistic and deterministic methodologies for the assessment of uncertainties of nuclear responses due to nuclear data uncertainties and their sensitivities to the involved reaction cross-section data. We have used MCNP and MCSEN codes in the Monte Carlo approach and DORT and SUSD3D in the deterministic approach for transport and sensitivity calculations, respectively. In both cases JEFF-3.1 and FENDL-2.1 libraries for the transport data and mainly ENDF/B-VI.8 and SCALE6.0 libraries for the relevant covariance data have been used. With a few exceptions, the two different methodological approaches were shown to provide consistent results. A total nuclear data related uncertainty in the range of 1–2% (1σ confidence level) was assessed for the tritium production in the HCLL mock-up experiment.