Yokoyama, Kenji; Sugino, Kazuteru; Ishikawa, Makoto; Maruyama, Shuhei; Nagaya, Yasunobu; Numata, Kazuyuki*; Jin, Tomoyuki*
JAEA-Research 2018-011, 556 Pages, 2019/03
We have developed a new unified cross-section set ADJ2017, which is an improved version of the unified cross-section set ADJ2010 for fast reactors. The unified cross-section set is used for reflecting information of C/E values (analysis / experiment values) obtained by integral experiment analyses; the values are stored in the standard database for FBR core design via the cross-section adjustment methodology, which integrates with the information such as uncertainty (covariance) of nuclear data, uncertainty of integral experiment / analysis, sensitivity of integral experiment with respect to nuclear data. The ADJ2017 is based on Japan's latest nuclear data library JENDL-4.0 as in the previous version of ADJ2010, and it incorporates more information on integral experimental data sets related to minor actinides (MAs) and degraded plutonium (Pu). In the creation of ADJ2010, a total of 643 integral experimental data sets were analyzed and evaluated, and 488 of the integral experimental data sets were finally selected to be used for the cross-section adjustment. In contrast, we have evaluated a total of 719 data sets, and eventually adopted 620 integral experimental data sets to create ADJ2017. ADJ2017 shows almost the same performance as ADJ2010 for the main neutronic characteristics of conventional sodium-cooled MOX-fuel fast reactors. In addition, for the neutronic characteristics related to MA and degraded Pu, ADJ2017 improves the C/E values of the integral experimental data sets, and reduces the uncertainty induced by the nuclear data. ADJ2017 is expected to be widely used in the analysis and design research of fast reactors. Moreover, it is expected that the integral experimental data sets used for ADJ2017 can be utilized as a standard database of FBR core design.
Takino, Kazuo; Sugino, Kazuteru; Yokoyama, Kenji; Jin, Tomoyuki*; Oki, Shigeo
Proceedings of 2018 International Congress on Advances in Nuclear Power Plants (ICAPP 2018) (CD-ROM), p.1214 - 1220, 2018/04
Yokoyama, Kenji; Jin, Tomoyuki; Hirai, Yasushi*; Hazama, Taira
JAEA-Data/Code 2015-009, 120 Pages, 2015/07
The second version of the versatile reactor analysis code system, MARBLE2, has been developed. A lot of new functions have been added inMARBLE2 by using the base technology developed in the first version (MARBLE1). Introducing the remaining functions of the conventional code system (JOINT-FR and SAGEP-FR), MARBLE2 enables one to execute almost all analysis functions of the conventional code system with the unified user interfaces of its subsystem, SCHEME. In particular, the sensitivity analysis functionality is available in MARBLE2. On the other hand, new built-in solvers have been developed, and existing ones have been upgraded. Furthermore, some other analysis codes and libraries developed in JAEA have been consolidated and prepared in SCHEME. In addition, several analysis codes developed in the other institutes have been additionally introduced as plug-in solvers. Consequently, -ray transport calculation and heating evaluation become available. As for another subsystem, ORPHEUS, various functionality updates and speed-up techniques have been applied based on user experience of MARBLE1 to enhance its usability.
Oizumi, Akito; Jin, Tomoyuki*; Ishikawa, Makoto; Kugo, Teruhiko
Annals of Nuclear Energy, 81, p.117 - 124, 2015/07
The uncertainty associated with physical quantities, such as nuclear data, needs to be quantitatively analyzed. The present paper illustrates an analysis methodology to investigate the physical mechanisms of burnup actinide composition with nuclear-data sensitivity based on the generalized depletion perturbation theory. The target in this paper is the MOX fuel of the light water reactor. We start with the discussion of the basic physical mechanisms for burnup actinide compositions using the reaction-rate flow chart on the burnup chain. After that, the physical mechanisms of the productions of Cm and Pu are analyzed in detail with burnup sensitivity calculation. Conclusively, we can identify the source of actinide productions and evaluate the indirect influence of the nuclear reactions if the physical mechanisms of burnup actinide composition are analyzed using the reaction-rate flow chart on the burnup chain and burnup sensitivity calculation. Finally, we demonstrate the usefulness of the burnup sensitivity coefficients in an application to determine the priority of accuracy improvement in nuclear data in combination with the covariance of the nuclear data. In addition, the target actinides and reactions are categorized into patterns according to a sensitivity trend.
Yokoyama, Kenji; Hazama, Taira; Numata, Kazuyuki*; Jin, Tomoyuki*
Annals of Nuclear Energy, 66, p.51 - 60, 2014/04
A comprehensive and versatile reactor analysis code system, MARBLE, has been developed. MARBLE is designed as a software development frame-work for reactor analysis, which offers reusable and extendible functions and data models based on physical concepts, rather than a reactor analysis code system. From a viewpoint of the code system, it provides a set of functionalities utilized in a detailed reactor analysis scheme for fast criticality assemblies and power reactors, and nuclear data related uncertainty quantication such as cross-section adjustment. MARBLE covers all phases required in fast reactor core design prediction and improvement procedures, i.e. integral experiment database management, nuclear data processing, fast criticality assembly analysis, uncertainty quantication, and power reactor analysis. In the present paper, these functionalities are summarized and system validation results are described.
Fukushima, Masahiro; Ishikawa, Makoto; Numata, Kazuyuki*; Jin, Tomoyuki*; Kugo, Teruhiko
Nuclear Data Sheets, 118, p.405 - 409, 2014/04
Oizumi, Akito; Jin, Tomoyuki*; Yokoyama, Kenji; Ishikawa, Makoto; Kugo, Teruhiko
JAEA-Data/Code 2013-019, 278 Pages, 2014/02
In design work for nuclear fuel cycle plants, decommissioning facilities and light water reactors (LWRs), it has been feasible to quantitatively evaluate the uncertainty of fuel burnup characteristics with identifying error sources arising from the analytical modeling or the related physical property such as nuclear data. Owing to the recent improvement of sensitivity analysis method and enhancement of computer capability, this new evaluation technology would be a promising strategy against the current demand for quality assurance, verification & validation (V&V) and accountability. The present report summarizes nuclear-data sensitivity of atomic number densities after burnup for the LWR fuels of UO and MOX in PWR and BWR. The analysis method is based on the generalized perturbation theory with JENDL-4.0 and a multi-purpose reactor analysis code MARBLE. The present study focuses on 35 fission products and 18 actinides. Sensitivities are calculated with respect to multigroup cross sections, half-lives and fission yields. Electronic files of the sensitivities are stored in a compact disk as a database. Important trends of the sensitivities are presented and their physical mechanisms are discussed. By incorporating the sensitivities with nuclear data covariance and post irradiation examination data, it would be possible to meet the demand for V&V and to break down the uncertainty due to nuclear data into dominant error sources. Thus, the sensitivities can be used to suggest the needs for nuclear data measurements and to extract those for reactor physics experiments in order to make the strategic deliberation of design rationalization.
Okumura, Keisuke; Sugino, Kazuteru; Kojima, Kensuke; Jin, Tomoyuki*; Okamoto, Tsutomu; Katakura, Junichi*
JAEA-Data/Code 2012-032, 148 Pages, 2013/03
A set of cross section libraries for the isotope generation and depletion calculation code ORIGEN2 was produced by using recent nuclear data JENDL-4.0. In this new library (ORLIBJ40), neutron-induced cross sections, fission product yields, isomeric ratios and half-lives were updated. ORLIBJ40 includes 24 libraries for typical UO or MOX fuels of PWR and BWR. In addition, it includes 36 libraries for various fast reactor fuels. ORLIBJ40 was applied to the post irradiation examination analyses of LWR nuclear spent fuels. As a result, it was confirmed that improvements were achieved especially for inventory and radioactivity estimations of minor actinides (Am and Cm isotopes) and fission products sensitive to cross sections (Eu and Sm isotopes) and for long-lived fission products (Se, etc.), compared with other existing ORIGEN2 libraries.
Sugino, Kazuteru; Ishikawa, Makoto; Numata, Kazuyuki*; Iwai, Takehiko*; Jin, Tomoyuki*; Nagaya, Yasunobu; Hazama, Taira; Chiba, Go*; Yokoyama, Kenji; Kugo, Teruhiko
JAEA-Research 2012-013, 411 Pages, 2012/07
Aiming at evaluating the core design prediction accuracy of fast reactors, various kinds of fast reactor core experiments/tests have been analyzed with the Japan's latest evaluated nuclear data library JENDL-4.0. Totally 643 characteristics of reactor physics experiments/tests and irradiation tests performed using the critical facilities: ZPPR, FCA, ZEBRA, BFS, MASURCA, ultra-small cores of LANL and power plants: SEFOR, Joyo, Monju were dealt. In analyses, a standard scheme/method for fast reactor cores was applied including detailed or precise calculations for best estimation. In addition, results of analyses were investigated from the viewpoints of uncertainties caused by experiment/test, analytical modeling and cross-section data in order to synthetically evaluate the consistency among different cores and characteristics. Further, by utilizing these evaluations, prediction accuracy of core characteristics were evaluated for fast power reactor cores that are under designing in the fast reactor cycle technology development (FaCT) project.
Sugino, Kazuteru; Jin, Tomoyuki*; Hazama, Taira; Numata, Kazuyuki*
JAEA-Data/Code 2011-017, 44 Pages, 2012/01
Fast reactor group constant sets UFLIB.J40 and JFS-3-J4.0 were prepared, which are based on the latest Japanese evaluated nuclear data library JENDL-4.0. Concerning UFLIB.J40, several fine group constant sets, which covered 70-group, 73-group, 175-group and 900-group structures, and the ultra fine group constant set were prepared. The number of nuclides for cross-sections of lumped fission products was extended so as to follow the extension of the number of fissile species for fission yield data.
Sugino, Kazuteru; Ishikawa, Makoto; Yokoyama, Kenji; Nagaya, Yasunobu; Chiba, Go; Hazama, Taira; Kugo, Teruhiko; Numata, Kazuyuki*; Iwai, Takehiko*; Jin, Tomoyuki*
Journal of the Korean Physical Society, 59(2), p.1357 - 1360, 2011/08
In order to improve the prediction accuracy of core performances in the fast reactor core design study, the unified cross-section set has been developed in Japan. The unified cross-section set, which combines a wide range of integral experimental information with differential nuclear data, is produced by using the cross-section adjustment technique based on the Bayesian parameter-estimation theory. A new set ADJ2010 is currently under development. The present paper describes the results of the cross-section adjustment for ADJ2010 which is based on the JENDL-4.0 data. The evaluation of the core design accuracy for a commercial power fast reactor core is also discussed. ADJ2010 will be released soon and will be expected to be utilized for core design of future fast reactors.
Yokoyama, Kenji; Tatsumi, Masahiro*; Hirai, Yasushi*; Hyodo, Hideaki*; Numata, Kazuyuki*; Iwai, Takehiko*; Jin, Tomoyuki*; Hazama, Taira; Nagaya, Yasunobu; Chiba, Go; et al.
JAEA-Data/Code 2010-030, 148 Pages, 2011/03
A next generation reactor analysis code system, MARBLE, has been developed. MARBLE is a successor of the fast reactor neutronics analysis code systems, JOINT-FR and SAGEP-FR (conventional system), which were developed for so-called JUPITER standard analysis methods. MARBLE has the equivalent analysis capability to the conventional system. On the other hand, burnup analysis functionality for power reactors as improved compared with the conventional system. In the development of MARBLE, the object oriented technology was adopted. As a result, MARBLE became an assembly of components for building an analysis code (i.e. framework) but not an independent analysis code system which simply receives input and returns output. Furthermore, MARBLE provides some pre-built analysis code systems such as the fast reactor neutronics analysis code system, SCHEME, which corresponds to the conventional code and the fast reactor burnup analysis code system, ORPHEUS.
Hirai, Yasushi*; Hyodo, Hideaki*; Tatsumi, Masahiro*; Jin, Tomoyuki*; Yokoyama, Kenji
JAEA-Data/Code 2008-020, 188 Pages, 2008/10
Japan Atomic Energy Agency promotes development of innovative analysis methods and models in fundamental studies for next-generation nuclear reactor systems. In order to efficiently and effectively reflect the latest analysis methods and models to primary design of prototype reactor and/or in-core fuel management for power reactors, a next-generation analysis system MARBLE has been developed. In this study, detailed design of a framework, its implementation and tests are conducted so that a Python system layer can drive calculation codes written in C++ and/or Fortran. It is confirmed that various type of calculation codes such as diffusion, transport and burnup codes can be treated in the same manner on the platform for unified management system for calculation codes with a data exchange mechanism for abstracted data model between the Python and the calculation code layers.
Yokoyama, Kenji; Jin, Tomoyuki*
Proceedings of International Conference on Nuclear Data for Science and Technology (ND 2007), Vol.2, p.807 - 810, 2008/05
Post irradiation experiment (PIE) data on depleted fuel composition of the experimental fast reactor "JOYO" MK-II has been accumulated since 1986 in JNC (the former of JAEA). In the present study, all the available PIE data of JOYO MK-II driver fuel were analyzed and integral data concerning U depletion and U generation were prepared. Both the integral data are sensitive to U capture cross section and applicable to nuclear data benchmarks. The recent evaluated nuclear data libraries, JENDL-3.2, -3.3, JEFF-3.1 and ENDF/B-VII, have a tendency to overestimate the generation of U. A cross section adjustment demonstrated that re-evaluation of U capture cross section improved the overestimation.
Oki, Shigeo; Jin, Tomoyuki*
Journal of Nuclear Science and Technology, 42(4), p.390 - 397, 2005/05
Burnup calculation and nuclear-data sensitivity analysis were carried out with respect to the heavy metal composition in fast reactors. The following nuclear data libraries were compared: JEF-2.2, ENDF/B-VI Release 5, JENDL-3.2, and JENDL-3.3. Differences for main heavy metal nuclides (235U, 238U, 239Pu, 240Pu, and 241Pu) were small since the number densities after 1-cycle burnup did not change over 1 or 2 % among those libraries. Relatively large differences were found for minor actinide nuclides, especially for 236U, 237Np, 242mAm, 243Am, and curium isotopes. The number densities for these nuclides after burnup showed remarkable library dependence over 5% through 50%. A burnup sensitivity analytical system based on the generalized perturbation theory enabled us to find out quantitatively the causative nuclides and reactions, as well as their energy regions.
Jin, Tomoyuki*; Oki, Shigeo
JNC-TN9410 2005-012, 54 Pages, 2005/03
For the design study of LLFP (Long-Lived Fission Product) transmutation core concepts in Feasibility Study on commercialized fast reactor cycle systems, correction factors for LLFP transmutation rate were evaluated by comparing the design calculation with the detailed Monte-Carlo calculation. By a step-by-step approach from design to detailed calculation, the correction factors were decomposed into the following factorial effects: (a) group collapsing effect, (b) transport effect, (c) difference between the deterministic method and the Monte-Carlo calculation, (d) detailed energy group effect (including a treatment of the thermal energy region), (e) three-dimensional effect of the core geometry, and (f) heterogeneity effect of the LLFP target assembly. As a result, the heterogeneity effect of the LLFP target assembly was turned out to be the main cause of the difference between the design and the detailed calculations. It is possible to explain the mechanism of the heterogeneity effect in connection with the cross-section shape of LLFP from resonance to thermal energy region.
Katakura, Junichi; Kataoka, Masaharu*; Suyama, Kenya; Jin, Tomoyuki*; Oki, Shigeo*
JAERI-Data/Code 2004-015, 115 Pages, 2004/11
A set of cross section libraries for ORIGEN2 code, ORLIBJ33, has been produced based on the latest Japanese Evaluated Nuclear Data Library JENDL-3.3. The produced libraries are for LWR's which include PWR, BWR and their MOX fuels. The libraries for FBR's are also produced. Using the libraries for LWR, comparisons with old libraries based on JENDL-3.2 were performed. The comparisons with measured PIE data were also carried out. For the libraries for FBR, the comparisons with the calculations using the old libraries were performed and the effects using different libraries were discussed.
Jin, Tomoyuki*; Oki, Shigeo
JNC-TN9410 2004-001, 79 Pages, 2004/03
ORLIBJ32 is a set of ORIGEN2 cross-section libraries for light water reactors and fast reactors based on Japanese evaluated nuclear data library JENDL-3.2. It has been opened since 1999. Following the latest revision of JENDL-3.2 to JENDL-3.3 in 2002, we have prepared new ORIGRN2 libraries for fast reactors. By using the same tool that made the fast reactor libraries in ORLIBJ32, the 73-group infinitely-diluted cross sections for 327 nuclides generated from JENDL-3.3 were collapsed into 1-group cross sections with an arbitrary weighting spectrum. For main nuclides, we used the 1-group shielded cross sections obtained by the fast reactor group constant set JFS-3-J3.3. As an isomeric ratio (g/(g+m)) for 241Am capture reaction, the value 0.85 was used instead of the conventional value of 0.80 in consequence of the latest research development of nuclear data. The new libraries were prepared for the following sodium-cooled fast reactors just like ORLIBJ32: JOYO (MK-I), MONJU, several kinds of a prototype reactor (600 MWe) parameterized by both fuel type (MOX, Metal, Nitride) and Pu isotopic composition, a commercial-size reactor (1300 MWe), and a Pu burning reactor. We performed burnup calculations with the new ORIGEN2 libraries in order to investigate the effect caused by the revision of the library.
Yokoyama, Kenji; Jin, Tomoyuki*
JNC-TN9400 2004-009, 64 Pages, 2004/02
In the development of fast reactors, it is important to verify and improve the prediction accuracy of isotopic composition change by burnup. In this study, post irradiation experiment (PIE) data of fuel assemblies irradiated in the experimental fast reactor "JOYO" MK-I core was analyzed. As the isotopic composition after irradiation had been measured in the PIE, the authors selected to evaluate the rate of change of atomic number density (RC) by burnup. The RCs strongly depend on the error of the initial atomic number density. Therefore, the authors evaluated RCs based on the sample regression constants with the linear-square method in order to remove the dependency on the error of the initial atomic number density. The analysis was carried out for all the avajlable - 80 point PIE data on core fuel In the analysis, the 70 group diffusion calculation with 3-dimensional Tri-Z geometry model was performed by using the nuclear data library JENDL-3.2, and the burn up calculation considering neutron spectrum in the irradiation position was carried out. As a result of the analysis, calculated value agreed with experimental values within 2--10 % on the main nuclides, such as U-235, U-236, U-238, Pu-239 and Pu-241. However, it is necessary to upgrade and verify the analysis method by using these PIE data since there is a room for improvement of the analysis models. 0n the other hand, the nuclear data library effect was also evaluated by using JENDL-3.3. It was found that JENDL-3.3 increased both RCs of U-236 and Pu-241 by -5% due to the revisions of U-235 and Pu-240 capture cross-sections. In the future, improvement of the analysis model and quantification of the error will make it possible to apply the PIE data of "JOYO" for the verification of these nuclear data.
Oki, Shigeo; Yokoyama, Kenji; Numata, Kazuyuki*; Jin, Tomoyuki*
JAERI-Conf 2004-005, p.40 - 45, 2004/00
To develop a commercialized fast reactor cycle system involving the recycling of minor actinide (MA) nuclides, Japan Nuclear Cycle Development Institute has launched the isotopic composition analysis of post irradiation examination (PIE) results for fuels and MA samples irradiated at the experimental fast reactor "JOYO". This paper presents the target accuracy of MA nuclear data, as well as the progress in validation of those by PIE analyses. The analysis result on the first examined MA sample suggested the necessity of re-evaluation of the isomeric ratio for 241Am capture reaction both in ENDF/B-VI and in JENDL-3.3.