研究者詳細 - 遠藤　知弘

MISC - 遠藤　知弘

分割表示 188 件中 61 - 80 件目／全件表示 >>

Application of Singular Value Decomposition and Low Rank Approximation for Compression of Macroscopic and Microscopic Cross Section Table for Core Calculations 査読有り国際誌

M. Yamamoto, T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 166 - 169 2019年12月

Uncertainty Quantification of Neutronics Characteristics in Thermal Systems Using Random Sampling and Continuous Energy Monte-Carlo Methods 査読有り国際誌

H. Oike, R. Kondo, T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 14 - 17 2019年12月

Performance of the RSE (Resonance calculation using energy Spectral Expansion) Method for Heterogeneous Pin-cell Geometry 査読有り国際誌

R. Kondo, T. Endo, A. Yamamoto, S. Takeda, H. Koike, K. Yamaji, D. Sato

Proceedings of Reactor Physics Asia 2019 頁： 162 - 165 2019年12月

Estimated Criticality Lower-Limit Multiplication Factor Considering Neutronic Similarity and Uncertainties of Effective Multiplication Factor Using the Bootstrap Method (2) Application 査読有り国際誌

F. Nishioka, T. Hayashi, T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 254 - 257 2019年12月

T. Hayashi, F. Nishioka, T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 250 - 253 2019年12月

Effectiveness of Subcritical Measurement at Solid Moderated KUCA Core for Reducing Nuclear Data-Induced Uncertainties in Other Light Water Reactor Analysis 査読有り国際誌

T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 262 - 265 2019年12月

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）

This paper aims to clarify that the subcritical measurement at the solid moderated KUCA core is useful to reduce nuclide data-induced uncertainties of the effective neutron multiplication factors keff in other light water moderated systems. For this purpose, we pay attention to the prompt neutron decay constant α, which was measured by the Feynman-α method with the moving block bootstrap method for the deep subcritical KUCA core under the shutdown state without any external neutron source. The sensitivity analyses of α for the subcritical KUCA core and of keff for various light water moderated critical cores were carried out based on the first-order perturbation theory. Consequently, it is demonstrated that the nuclear data-induced uncertainties of keff can be reduced by the bias factor method using the measurement value of α, if the target keff value has a strong correlation with the measured α value in the subcritical KUCA core.

Discontinuity Factor; A Discontinuity Condition for Angular Flux? 査読有り国際誌

A. Yamamoto, T. Endo

Proceedings of Reactor Physics Asia 2019 頁： 155 - 157 2019年12月

C5G7-TD Benchmark Analysis using Multigrid Amplitude Function Method 査読有り国際誌

K. Tsujita, T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 201 - 204 2019年12月

Application of Surrogate Modeling with Singular Value Decomposition for Design Basis Accident Aiming Statistical Safety Analysis 査読有り国際誌

M. Matsushita, T. Endo, A. Yamamoto

Proceedings of Reactor Physics Asia 2019 頁： 282 - 285 2019年12月

Development of Efficient Data Sampling Method to Construct Surrogate Model of Severe Accident Analysis Code for SBO Aiming Probabilistic Safety Margin Analysis 査読有り国際誌

M. Matsushita, T. Endo, A. Yamamoto

Trans. Am. Nucl. Soc.121 巻頁： 979 - 982 2019年11月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

An efficient data sampling method is developed in order to construct a ROM, which accurately reproduces the results of a severe accident code, with the small number of the training data. The proposed sampling method is categorized as the adaptive sampling and this study newly applies the concept of the gravity field to its sampling method in order to make efficient sampling. The proposed sampling method can choose the input training data which have significant new information for ROM construction. The proposed sampling method is applied to PCT obtained by RELAP5/SCDAPSIM calculations for the SBO accident scenario. ROM is improved by adding the new training data to satisfy MPCT=1200°C on the core intact/damage boundary. The core intact/damage boundary predicted by ROM improved as the number of iterations increases. In other words, the input training data are heavily sampled near the core intact/damage boundary. This sampling makes the proposed method more efficient than the conventional random sampling method. As future problems, applications of the present method to calculate CDF considering the uncertainty of the ROM and to the other accident scenarios will be considered.

DOI： 10.13182/T30744

Resonance Calculation Using Energy Spectral Expansion Based on Reduced Order Model: Application to Heterogeneous Geometry 査読有り国際誌

A. Yamamoto, R. Kondo, T. Endo, S. Takeda, H. Koike, K. Yamaji, D. Sato

Trans. Am. Nucl. Soc.121 巻頁： 1316 - 1320 2019年11月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

The Resonance calculation based on energy Spectral Expansion (RSE) is applied to a pin cell geometry and its accuracy is verified through comparison with the results by the ultra-fine group MOC and the MCNP calculations. Impact of the generation method of the orthogonal bases and the neutron source treatment is quantified through the benchmark calculation. The numerical results are promising; the RSE method accurately reproduces the reference results obtained by the ultra-fine group calculation by MOC or MCNP. There are several issues to be addressed; performance of the present method in more complicated and practical geometries/material arrangements, treatment of scattering source term, numerical solution method, extension of the present concept to the unresolved resonance energy range.

DOI： 10.13182/T30993

Reduction of Macroscopic and Microscopic Cross Section Table Size for Heterogeneous Core Calculation using Dimensionality Reduction 査読有り国際誌

M. Yamamoto, T. Endo, A. Yamamoto

Trans. Am. Nucl. Soc.121 巻頁： 1309 - 1312 2019年11月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

Reduction of cross section table size including macroscopic and microscopic cross sections is tried by applying the dimensionality reduction technique with the singular value decomposition and the low-rank approximation. Among the material-wise cross sections obtained by a lattice physics calculation, those of fuel region are considered in this study. The impact of dimensionality reduction is investigated with a comparison of the assembly calculations using the reconstructed and original cross sections. In this study, approximately 99% reduction in cross section table size is possible while suppressing the error of keff and flux below 0.1%. As a future study, development of the tabulation method using dimensionality reduced cross sections is considered to verify the applicability of the present method for core calculations.

DOI： 10.13182/T30783

基礎から分かる未臨界：(第1回) 今なぜ未臨界か査読有り

遠藤知弘, 辻本和文, 山本章夫

アトモス : 日本原子力学会誌61 巻 ( 10 ) 頁： 734 - 738 2019年10月

担当区分：筆頭著者記述言語：日本語掲載種別：記事・総説・解説・論説等（商業誌、新聞、ウェブメディア）出版者・発行元：日本原子力学会

臨界管理や原子炉の運転等において，体系の中性子実効増倍率は最も基本的な概念の一つである。実効増倍率が1のときを臨界，1未満のときを未臨界と呼ぶことは良く知られているが，実は両者は大きく異なる状態である。本連載講座では，核燃料サイクル施設の臨界安全，原子炉施設の臨界管理や福島第一原子力発電所のデブリ取り出し等で重要となる未臨界状態について，臨界状態の原子炉との違い，炉物理的な特徴，未臨界状態であることを評価・測定する手法などを解説する。第1回では，臨界・未臨界の炉物理的な違いについて紹介する。

DOI： 10.3327/jaesjb.61.10_734

Conversion from prompt neutron decay constant to subcriticality using point kinetics parameters based on α- and keff-eigenfunctions 査読有り国際誌

T. Endo, A. Yamamoto

Proc. ICNC2019 2019年9月

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）

The conversion from the measurement value of prompt neutron decay constant α to the subcriticality –ρ is investigated, using special point kinetics parameters which are defined by mixing forward α- and adjoint keff-eigenfunctions. The mixing α-k weighted point kinetics parameters are very useful to reduce the conversion bias of neutron multiplication factor in a deeper subcritical system. In the same manner as the k-ratio method, the mixing α-k weighted point kinetics parameters can be well approximated by three eigenvalues of α, keff, and prompt keff,p to be simply estimated by the forward eigenvalue calculations without any adjoint calculations.

Measurement of subcriticality in dollar units using time-domain decomposition based integral method 査読有り国際誌

A. Nonaka, T. Endo, A. Yamamoto, M. Yamanaka, T. Sano, C.H. Pyeon

Proc. ICNC2019 2019年9月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）

This paper presents an estimation method of subcriticality in dollar units developed on the basis of an integral method for arbitrary state changes in a subcritical system. In a general transient in a subcritical system, reactivity, neutron source intensity, and point kinetics parameters (and) can vary simultaneously. To address this problem, the "time-domain decomposition based integral method (TDDI)" has been proposed. The TDDI method can estimate the subcriticality only using the time variation of the neutron count rate. Therefore, the proposed method is useful to estimate the subcriticality in the reactor where the detailed conditions are unknown. To investigate the applicability of the TDDI method to actual subcritical measurement, a transient experiment in a source-driven subcritical system is conducted at the Kyoto University Critical Assembly. As a result, it is concluded that the TDDI method can approximately estimate the order of magnitude for the subcriticality in dollar units. Meanwhile, the estimation is difficult owing to the statistical errors when the neutron count rate is low.

Application of Various Superhomogenization (SPH) methods for the Method of Characteristics 査読有り国際誌

K. Sawada, T. Endo, A. Yamamoto

Proc. M&C2019 頁： 1534 - 1543 2019年8月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

To reduce the error due to the difference between the fine and the coarse calculation conditions in a transport calculation using method of characteristics (MOC), the superhomogenization (SPH) method has been widely used. However, the conventional SPH method has a problem with numerical stability in the iterative calculations for the estimation of SPH factors, especially when local strong absorbers exist in a calculation geometry. To address this problem, improvements of the SPH methods were tested to reduce the angular and spatial discretization error of the MOC. The performances of the various SPH methods were confirmed in pressurized water reactor (PWR) 2 × 2 fuel assembly geometries and in a PWR core geometry. The results of verification indicate that in a core geometry the improved SPH methods provide better accuracy than that of the conventional SPH.

Implementation of Random Sampling for ACE-format Cross Sections using FRENDY and Application to Uncertainty Reduction 査読有り国際誌

R. Kondo, T. Endo, A. Yamamoto, K. Tada

Proc. M&C2019 頁： 1493 - 1502 2019年8月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

For uncertainty quantification and reduction using the random sampling technique through a continuous energy Monte Carlo method, the perturbation capability of nuclear data libraries for MCNP is developed using a nuclear data processing system FRENDY, which is being developed by JAEA. The implemented capability is applied to uncertainty quantification using the random sampling method and to uncertainty reduction of kinetic parameter based on the bias factor method. Validity of the implemented capability is confirmed through comparison with the results obtained by the conventional sandwich formula using SCALE and MCNP.

Calculation Method of Estimated Criticality Lower-limit Multiplication Factor using the Bootstrap Method 査読有り国際誌

T. Hayashi, T. Endo, A. Yamamoto

Proc. M&C2019 頁： 1874 - 1885 2019年8月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

The present paper describes a calculation method of the estimated criticality lower-limit multiplication factor (ECLLMF) using the bootstrap method, which is an uncertainty evaluation method. In the conventional method, the normal distribution is assumed for a probability distribution with respect to calculation biases of effective multiplication factors (keff) to evaluate ECLLMF. To propose a methodology without the assumption of the normal distribution, an estimation method of ECLLMF is newly developed using the bootstrap method, where the assumption of the normal distribution is not necessary. From the verification results, it is confirmed that the estimation result using the non-parametric bootstrap method is more reasonable than that using the conventional method, when a probability distribution of keff does not obey the normal distribution.

A Resonance Calculation Method using Energy Expansion Bases based on a Reduced Order Model 査読有り国際誌

A. Yamamoto, T. Endo, S. Takeda, H. Koike, K. Yamaji, K. Ieyama, D. Sato

Proc. M&C2019 頁： 1081 - 1092 2019年8月

記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

A Resonance calculation using energy Spectral Expansion bases (the RSE method) is newly proposed. In the present method, an ultra-fine group spectrum appeared in a resonance calculation is expanded by expansion bases on energy. The transport equation for the expansion coefficients is derived, which is the simultaneous first order differential equation for the expansion coefficients. The present method is applied to a one-dimensional slab geometry composed of U238. Comparisons of the results obtained by the present and the ultra-fine group (reference) methods show the fundamental validity of this method.

Data Assimilation using Subcritical Measurement of Prompt Neutron Decay Constant 査読有り国際誌

T. Endo, A. Yamamoto

Proc. M&C2019 頁： 1864 - 1873 2019年8月

担当区分：筆頭著者,　責任著者記述言語：英語掲載種別：研究発表ペーパー・要旨（国際会議）出版者・発行元：American Nuclear Society

The prompt neutron decay constant α can be directly measured by the reactor noise analysis method (e.g., the Feynman-α method) in a steady-state subcritical system. In this study, the applicability of the data assimilation techniques (i.e., the bias factor and cross section adjustment methods) using a subcritical measurement of α conducted at the Kyoto University Critical Assembly (KUCA) was investigated to reduce the nuclear data-induced uncertainty of k_eff.The sensitivity coefficients of k_eff and α with respect to the nuclear data were efficiently estimated using a deterministic SN transport code with the first-order perturbation theory. As a result, a priori relative uncertainty of k_eff due to the 56 group SCALE covariance data can be reduced. The experimental value of α contributes to improving nuclear data of fission spectrum χ and total fission neutron number ν via strong correlations between χ and prompt χ_p and between ν and prompt ν_p.

　前のページ - 次のページ　

<遠藤 知弘>

MISC - 遠藤 知弘

<遠藤　知弘>

MISC - 遠藤　知弘