研究者詳細 - 山本　章夫

論文 - 山本　章夫

分割表示 475 件中 141 - 160 件目／全件表示 >>

Estimation of sensitivity coefficient based on lasso-type penalized linear regression

Katano, R; Endo, T; Yamamoto, A; Tsujimoto, K

JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY 55 巻 ( 10 ) 頁： 1099 - 1109 2018年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Journal of Nuclear Science and Technology

We proposed the penalized regression ‘adaptive smooth-lasso’ for the estimation of sensitivity coefficients of the neutronics parameters. The proposed method utilizes the variation of the microscopic cross-sections and the neutronics parameters obtained by random sampling. The weighted penalty term of the proposed method is more appropriate for the estimation of the sensitivity of neutronics parameters to the microscopic cross-section than that of the conventional methods. In a numerical verification calculation, sensitivity coefficients of keff of an accelerator-driven system are estimated using the proposed method, the conventional penalized regression, and the direct method. Comparison of these results indicates that the proposed method is superior to the conventional penalized linear regression from the viewpoint of reproduction of the reference sensitivity coefficients obtained by the direct method. Through the verification calculations, the proposed method can be a candidate for a practical method to estimate the sensitivity coefficients with low calculation cost.

DOI： 10.1080/00223131.2018.1479988

Web of Science

Scopus

Radially and azimuthally dependent resonance self-shielding treatment for general multi-region geometry based on a unified theory

Koike, H; Kirimura, K; Yamaji, K; Kosaka, S; Yamamoto, A

JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY 55 巻 ( 1 ) 頁： 41 - 65 2018年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Journal of Nuclear Science and Technology

A unified resonance self-shielding method, which can treat general sub-divided fuel regions, is developed for lattice physics calculations in reactor physics field. In a past study, a hybrid resonance treatment has been developed by theoretically integrating equivalence theory and ultra-fine-group slowing-down calculation. It can be applied to a wide range of neutron spectrum conditions including low moderator density ranges in severe accident states, as long as each fuel region is not sub-divided. In order to extend the method for radially and azimuthally sub-divided multi-region geometry, a new resonance treatment is established by incorporating the essence of sub-group method. The present method is composed of two-step flux calculation, i.e. ‘coarse geometry + fine energy’ (first step) and ‘fine geometry + coarse energy’ (second step) calculations. The first step corresponds to a hybrid model of the equivalence theory and the ultra-fine-group calculation, and the second step corresponds to the sub-group method. From the verification results, effective cross-sections by the new method show good agreement with the continuous energy Monte-Carlo results for various multi-region geometries including non-uniform fuel compositions and temperature distributions. The present method can accurately generate effective cross-sections with short computation time in general lattice physics calculations.

DOI： 10.1080/00223131.2017.1384704

Web of Science

Scopus

Dimension-reduced cross-section adjustment method based on minimum variance unbiased estimation

Yokoyama, K; Yamamoto, A; Kitada, T

JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY 55 巻 ( 3 ) 頁： 319 - 334 2018年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Journal of Nuclear Science and Technology

A new formulation of the cross-section adjustment methodology with the dimensionality-reduction technique has been derived in the light of the fact that it is often used under the condition of ill-posed problem, where the number of integral experimental quantities is less than the number of adjusted nuclear data parameters. This new formulation is proposed as the dimension-reduced conventional cross-section adjustment method (DRCA). The derivation of DRCA is based on the minimum variance unbiased estimation (MVUE), and the assumption of normal distribution is not used. The result of DRCA depends on a user-defined matrix that determines the dimension-reduced feature subspace. We examined three variations of DRCA, namely, DRCA1, DRCA2, and DRCA3, which employ (1) the nuclear data covariance matrix as the user-defined matrix, (2) the sensitivity coefficient matrix postmultiplied by the nuclear data covariance matrix, and (3) the sensitivity coefficient matrix, respectively. Mathematical investigation and numerical verification revealed that DRCA2 is equivalent to the currently widely used cross-section adjustment method. Moreover, DRCA3 is found to be identical to the cross-section adjustment method based on MVUE, which has been proposed in the previous study.

DOI： 10.1080/00223131.2017.1397563

Web of Science

Scopus

Estimation of Sensitivity Coefficient based on Lasso-type Penalized Linear Regression 査読有り

R. Katano, T. Endo, A. Yamamoto, K. Tsujimoto

JOURNAL OF NUCLEAR SCIENCE AND TECHNOLOGY 55 巻頁： 1099-1109 2018年

Utilization of Region-wise Even-Parity Discontinuity Factor to Reduce Discretization Error of MOC 査読有り

A. Yamamoto, A. Giho, T. Endo

Nuclear Science and Engineering 193 巻頁： 253-268 2018年

Flux Region Assignment Method using Ray Trace Information for the Method of Characteristics to Improve Cache Efficiency

A. Yamamoto, A. Giho, T. Endo

Nuclear Science and Engineering 192 巻頁： 243-250 2018年

Flux Region Assignment Method Using Ray Trace Information for the Method of Characteristics to Improve Cache Efficiency

Yamamoto, A; Giho, A; Endo, T

NUCLEAR SCIENCE AND ENGINEERING 192 巻 ( 3 ) 頁： 240 - 253 2018年

記述言語：英語掲載種別：研究論文（学術雑誌）出版者・発行元：Nuclear Science and Engineering

A flux region assignment algorithm to increase cache efficiency for the method of characteristics (MOC) is proposed. In order to minimize the stride of memory access, flux region identifications are assigned based on the ray trace sequence during the MOC calculation. The present method is implemented in the three-dimensional transport code GENESIS and its performance is confirmed through verification calculations ranging from single pressurized water reactor (PWR) fuel assembly to PWR full-core benchmark problems. Quantitative comparison of cache efficiency is carried out and the present method shows improved cache efficiency, which results in a reduction in computation time. The present method can reduce computational time by improving cache efficiency while suppressing memory requirement.

DOI： 10.1080/00295639.2018.1501978

Web of Science

Scopus

リスク評価とマネジメントに関するアジアシンポジウム開催報告

山口彰, 山本章夫, 成宮祥介

日本原子力学会誌ATOMOΣ 60 巻 ( 6 ) 頁： 362 - 363 2018年

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：一般社団法人日本原子力学会

<p>　リスクに関する国際シンポジウムは，日中韓を中心に横浜で開催された。20年ほど前に日韓で始まった日韓PSAワークショップにその起源をおくもので，原子力施設に関するリスク評価の専門家が一堂に会し，充実した意見交換，議論が行われた。リスクの本格活用に臨む我が国としては多くの知見を得たシンポジウムとなった。</p>

DOI： 10.3327/jaesjb.60.6_362

CiNii Research

原子炉物理分野の研究開発ロードマップ2017:次世代が考える炉物理の未来

山本章夫, 千葉豪, 桐村一生, 三木陽介, 横山賢治

日本原子力学会誌ATOMOΣ 60 巻 ( 4 ) 頁： 241-245 2018年

記述言語：英語掲載種別：研究論文（学術雑誌）

<p>　炉物理部会の傘下に設置された「炉物理ロードマップ調査・検討」WGにおけるロードマップ策定の概要を紹介する。本ロードマップの特徴は，①次世代を担う若手の技術者・研究者を中心に議論・策定を進めたこと，②現状から類推して課題を設定するフォアキャストアプローチに加え，原子炉物理分野のビジョンとミッションを検討し，これらを達成するために解決すべき課題をバックキャストアプローチにより設定したこと，にある。本ロードマップの詳細は，報告書として炉物理部会のホームページより閲覧可能である。</p>

DOI： 10.3327/jaesjb.60.4_241

Surrogate model of severe accident analysis code for SBO aiming probabilistic safety margin analysis

Matsushita M., Endo T., Yamamoto A.

Transactions of the American Nuclear Society 119 巻頁： 900 - 903 2018年

Sensitivity coefficient analysis of omega-eigenvalue based on first-order perturbation theory

Endo T., Yamamoto A.

International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems Part F168384-2 巻頁： 1240 - 1253 2018年

記述言語：日本語掲載種別：研究論文（学術雑誌）出版者・発行元：International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems

Experimental results of prompt neutron decay constant α is useful information to validate numerical results of ω -eigenvalue for spatial and energetic fundamental mode. In order to accomplish the data assimilation technique using α, it is desirable to establish an efficient numerical calculation method for sensitivity coefficient analysis of ω -eigenvalue. For this purpose, the numerical calculation method using the standard first-order perturbation theory is investigated. A specific theoretical formula is derived to evaluate the sensitivity coefficient of ω to nuclear data. The derived formula utilizes forward and adjoint eigenfunctions which consist of neutron flux and delayed neutron precursor densities. Through a feasibility study based on the multi-group diffusion calculation, the derived formula is verified by comparing with the reference results using the direct method. In addition, it is confirmed that the prompt approximation is applicable to the evaluation of sensitivity coefficient of α for a subcritical state where α is sufficiently larger than decay constants of delayed neutron precursors.

Scopus

Reduction of cross section table size for core analysis using dimensionality reduction technique

Yamamoto M., Endo T., Yamamoto A.

Transactions of the American Nuclear Society 119 巻頁： 1226 - 1228 2018年

Cache efficient flux region assignment for the method of characteristics

Yamamoto A., Giho A., Endo T.

International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems Part F168384-2 巻頁： 788 - 799 2018年

A flux region assignment algorithm to increase cache efficiency for the method of characteristics is proposed. In order to minimize the stride of memory access, flux region IDs are assigned based on the ray trace sequence during the method of characteristics calculation. The present method is implemented in the three-dimensional transport code GENESIS and its performance is confirmed through verification calculations ranging from single PWR fuel assembly to PWR full core benchmark problems. The present method can reduce computational time by improving cache efficiency while suppressing memory requirement.

Scopus

Development of reduced order model of severe accident analysis code for probabilistic safety margin analysis

Matsushita M., Endo T., Yamamoto A., Kitao T.

International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems Part F168384-5 巻頁： 3042 - 3053 2018年

For probabilistic safety margin analysis, we developed a reduced order model (ROM) that reconstruct calculation results of typical severe accident progressions obtained by the severe accident analysis code, MAAP. The ROM is applied for fast reconstruction of time series plant parameters (e.g., pressure and temperature) obtained by the MAAP code. The ROM is applied to two severe accident scenarios, i.e., the station black out (SBO) with loss of all feed water capabilities and the large break loss of coolant accident (large break LOCA) without the emergency core cooling system (ECCS) capability. Verification results indicate that the ROM reasonably reproduces temporal variation of plant parameters with a few bases obtained by the ROM, which enables very fast reconstruction of complicated accident progression of a severe accident.

Scopus

Development of the uncertainty quantification method of activation in reactor structures using reduced-order modeling

Yokoi K., Endo T., Yamamoto A., Hayashi K., Mizuno R., Kimura Y.

International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems Part F168384-3 巻頁： 1793 - 1804 2018年

Uncertainty quantification of activation in structure materials around a nuclear reactor is important for efficient planning of decommissioning. In order to quantify the uncertainty of activation in reactor structures, neutron shielding calculations are necessary. Since the neutron shielding calculation is performed for spatially large regions around a reactor, it requires a lot of computation time especially in the case of multi-dimensional geometry. In this study, the reduced-order modeling (ROM) is applied to the sensitivity and uncertainty analysis of activation in reactor structures to reduce computation time. The ROM can reduce calculation cost for evaluation of sensitivity coefficients by identifying important or sensitive subspace (active subspace, AS) in input data. In this study, an AS is constructed by the several sensitivity coefficients in reactor structures which are evaluated by the perturbation theory (PT). Sensitivity coefficients of activation throughout reactor structures are estimated using the AS and the uncertainty of activation is evaluated by the "sandwich formula." The calculation results indicate that the uncertainty of the activation in reactor structures can be reproduced with low calculation cost (approximately 30 neutron transport calculations) using the ROM in one-dimensional geometry of a 500 MWe class BWR.

Scopus

Estimation of region-wise even-parity discontinuity factor for MOC through iterative procedure

Yamamoto A., Giho A., Endo T.

International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems Part F168384-3 巻頁： 1892 - 1903 2018年

To reduce angular and spatial discretization error of MOC with a coarse calculation condition, region-wise even-parity discontinuity factor (EPDF) for transport calculations is evaluated through an iterative procedure using only region-wise scalar flux, i.e., without odd parity angular flux, partial-, or net-current at region boundary. Region-wise EPDF is evaluated in a single assembly geometry with reflective boundary condition. The evaluated EPDFs are applied to a 2x2 colorset assembly configuration and its performance is compared to the conventional superhomogenization (SPH) method. The calculation results indicate that 1) no convergence issue is observed during the iteration process to estimate EPDF, 2) performance of the region-wise EPDF is better than that of the conventional SPH method.

Scopus

Estimation of subcriticality in dollar units based on integral method for arbitrary state-change in subcritical system

Nonaka A., Endo T., Yamamoto A.

Transactions of the American Nuclear Society 119 巻頁： 1112 - 1115 2018年

Estimation of subcriticality in dollar units using integral method for subcritical system

Nonaka A., Endo T., Yamamoto A.

International Conference on Physics of Reactors, PHYSOR 2018: Reactor Physics Paving the Way Towards More Efficient Systems Part F168384-5 巻頁： 3271 - 3282 2018年

A subcriticality measurement method using the integral method is developed for any stepwise transient. The integral method usually applied to a negative reactivity insertion using the rod-drop and the source-jerk methods. In this study, it is clarified that the integral method can be applied to measure positive reactivity insertion. During refueling of a reactor, not only reactivity but also neutron source intensity and neutron generation time simultaneously change. In order to accurately monitor the subcriticality under these conditions, variations of these parameters must be taken into consideration. Therefore, a subcriticality measurement method is developed, which is applicable to not only variation of reactivity but also simultaneous variations of neutron source intensity and neutron generation time. Furthermore, the present method utilizes only measurement value of neutron count rate, thus the proposed method is practical. Verification calculations are carried out for a step-wise change of reactivity, neutron source intensity, and neutron generation time. The results indicate that reactivity is accurately predicted for the change of the neutron source intensity, which is difficult to achieve by the neutron source multiplication method. On the other hand, although the change of the neutron generation time has considerable impacts on the estimated result of subcriticality, the predicted error is less than 10% for very large variation of neutron generation time.

Scopus

Estimation of subcriticality using particle filter method

Ikeda T., Kimura T., Endo T., Yamamoto A.

Transactions of the American Nuclear Society 118 巻頁： 851 - 854 2018年

Estimation of subcriticality using particle filter method

Ikeda T., Kimura T., Endo T., Yamamoto A.

AISTech - Iron and Steel Technology Conference Proceedings 2018-May 巻頁： 851 - 854 2018年

　前のページ - 次のページ　

<山本 章夫>

論文 - 山本 章夫

<山本　章夫>

論文 - 山本　章夫