Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Katano, Ryota; Yamamoto, Akio*; Endo, Tomohiro*
Nuclear Science and Engineering, 196(10), p.1194 - 1208, 2022/10
Times Cited Count：1 Percentile：31.61(Nuclear Science & Technology)In this study, we propose the ROM-Lasso method that enables efficient evaluation of sensitivity coefficients of neutronics parameters to cross-sections. In the proposed method, a vector of sensitivity coefficients is expanded by subspace bases, so-called Active Subspace (AS) based on the idea of Reduced Order Modeling (ROM). Then, the expansion coefficients are evaluated by the Lasso linear regression between cross-sections and neutronics parameters obtained by the random sampling. The proposed method can be applied in the case where the adjoint method is difficult to be applied since the proposed method uses only forward calculations. In addition, AS is an effective subspace that can expand the vector of sensitivity coefficients with the lower number of dimension. Thus, the number of unknows is reduced from the original number of input parameters and the calculation cost is dramatically improved compared to the Lasso regression without AS. In this paper, we conducted ADS burnup calculations as a verification. We have shown how AS bases are obtained and the applicability of the proposed method.
Katano, Ryota; Yamamoto, Akio*; Endo, Tomohiro*
Proceedings of International Conference on Physics of Reactors 2022 (PHYSOR 2022) (Internet), p.2032 - 2041, 2022/05
We have proposed the ROM-Lasso method to perform an efficient evaluation of the sensitivity coefficients of ADS core parameters to cross sections without major modification of the core analysis system. In the ROM-Lasso method, the sensitivity coefficient vector is expanded via the subspace bases so-called Active Subspace (AS), and the effective number of unknowns is reduced. Then, the expansion coefficients are determined via the penalized linear regression with the core parameters obtained by the random sampling, and the sensitivity coefficient vector is estimated. Owing to the AS, the required number of the core calculations is dramatically reduced in the ROM-Lasso method. In this work, we take the sensitivity coefficient evaluation of the coolant void reactivity at the end of the cycle for example and demonstrate how estimation accuracy depends on the number of samples and the AS.
Iwamoto, Hiroki; Meigo, Shinichiro
Journal of Nuclear Science and Technology, 57(3), p.276 - 290, 2020/03
Times Cited Count：2 Percentile：24.28(Nuclear Science & Technology)This paper presents an approach to uncertainty estimation of spallation particle multiplicity of lead (
Pb), primarily focusing on proton-induced spallation neutron multiplicity (
) and its propagation to a neutron energy spectrum. The uncertainty is estimated from experimental proton-induced neutron-production double-differential cross sections (DDXs) and model calculations with the Particle and Heavy Ion Transport code System (PHITS). Uncertainties in multiplicities for 
, 
, and 
reactions are then inferred from the estimated uncertainty and the PHITS calculation. Using these uncertainties, uncertainty in a neutron energy spectrum produced from a thick Pb target bombarded with 500 MeV proton beams, measured in a previous experiment, is quantified by a random sampling technique, and propagation to the neutron energy spectrum is examined. Relatively large uncertainty intervals (UIs) were observed outside the lower limit of the measurement range, which is prominent in the backward directions. Our findings suggest that a reliable assessment of spallation neutron energy spectra requires systematic DDX experiments for detector angles and incident energies below 100 MeV as well as neutron energy spectrum measurements at lower energies below 
1.4 MeV with an accuracy below the quantified UIs.
Katano, Ryota; Endo, Tomohiro*; Yamamoto, Akio*; Tsujimoto, Kazufumi
Journal of Nuclear Science and Technology, 55(10), p.1099 - 1109, 2018/10
Times Cited Count：3 Percentile：30.05(Nuclear Science & Technology)In this study, we propose the penalized regression "adaptive smooth-lasso" for the estimation of sensitivity coefficients of the neutronics parameters. The proposed method estimates the sensitivity coefficients of the neutronics parameters using the variation of the microscopic cross sections and the neutronics parameter obtained by random sampling. The proposed method utilizes only the forward calculations. Thus, the proposed method can be applied for the complex reactor analysis for which the application of the adjoint method is difficult. In this study, we proposed a penalty term considering the characteristics of the sensitivity coefficients of the neutronics parameter to the microscopic multi-group cross sections. Through verification calculation, we show that the proposed method achieves high accuracy with less computational cost compared to the method based on random sampling proposed in the previous studies.
