Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Nagaya, Yasunobu; Mori, Takamasa
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
Nuclear reactor analysis requires calculations of reactivity worth such as control rod worth, void reactivity worth, sample reactivity worth, etc. It is, however, difficult to perform such calculations with the Monte Carlo method if the worth is small. In the present work, we extend the differential operator sampling method such that the derivative terms up to the fourth-order and the perturbed source effect up to the fourth-order can be estimated, and examine the applicability of the fourth-order differential operator sampling method to reactivity worth calculations. As a benchmark calculation, we perform the reactivity worth calculation for Godiva. The perturbation is introduced by decreasing the density in the central region of a radius of 1 cm. The result with the fourth-order differential operator sampling method agrees well with the reference one. As a practical calculation, we also perform Np sample worth calculation for TCA. The calculated sample reactivity worth with the differential sampling method approaches to the reference value as the higher-order effect is taken into account up to the fourth order. However, there still exists a discrepancy of
17%. It is, thus, found that many more higher-order effects must be taken into account or another method must be applied in this case.
Tatekawa, Takayuki; Teshima, Naoya; Suzuki, Yoshio; Takemiya, Hiroshi
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
In the nuclear field, various large-scale integrated simulations which cannot be executed with single job have been developed to reveal physical and engineering phenomena. Such integrated simulations are accomplished by coupling several simulation codes, each of which is charge of each physical process or each engineering part of whole system. Fault-tolerant (FT) mechanism is very important to run such simulations on the error-prone environment such as Grid. We developed functions of error detection, job re-submission, and file re-transfer and integrated them as a FT mechanism. Our test run of integrated nuclear energy application showed that the FT mechanism sustained the long run of the application by recovering the job failure automatically.
Kim, G.; Nakajima, Kohei*; Teshima, Naoya; Tatekawa, Takayuki; Suzuki, Yoshio; Takemiya, Hiroshi
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 5 Pages, 2010/10
Full-Scale 3D vibration simulator for an entire nuclear power plant is a seismic response analysis system for a whole digitalized nuclear power plant. In the system, boundary data from large components are used as input data of small components. To make a whole simulation efficient, we introduced pipeline method in which the data were transferred each time step running all components simulations in parallel. In the realization of the method on grid, since there were no existing grid technologies to sufficiently support the method for a long time, we developed simple orchestration application framework (SOAF) and using the SOAF, we performed seismic response analysis of a test reactor of JAEA and succeeded simulation for a week.
Nagatake, Taku; Kunugi, Tomoaki*; Takase, Kazuyuki
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 5 Pages, 2010/10
In the nuclear engineering fields, a high performance computer system is necessary to perform the large scale computations. Recently, a GPU (Graphics Processing Unit) has been developed as a rendering computational system in order to reduce a CPU (Central Processing Unit) load. The GPU consists of many processing units and a wide memory bandwidth in order to achieve the high-performance computing for rendering the high quality 3D objects. Nowadays, the performance of GPU has become much higher than that of the CPU, so the attempt to make use of the GPU for the scientific computations and more broad purposes is being performed. In this study, the MARS (Multi-interfaces Advection and Reconstruction Solver) which is one of the interface volume tracking methods for multiphase flows was developed by using the GPU.
Kino, Chiaki; Tatekawa, Takayuki; Teshima, Naoya; Kim, G.; Suzuki, Yoshio; Araya, Fumimasa; Nishida, Akemi; Takemiya, Hiroshi
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
In the present study, we have developed a new control system for application integration with the fault-tolerant API (FT-API). The system treats an application as a task which consists of one job and multiple file transfer. Firstly, each task designates a computer to submit a job using a scheduler associated to the job. Secondly, all files which are necessary to execute the job are gathered in the computer using FT-API for file transfer. Finally, the job is submitted using FT-API for job execution. If the computer is outage, the task designates a new computer, gathers necessary files and submits a new job. Each scheduler, file transfer and job condition can be flexibly defined in XML. This time, we applied the system to fluid-structure interaction analysis simulation. The result indicates that the system enables a user to easily execute multi-scale and multi-physics simulation using application integration.
Yamada, Tomonori
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
CCES/JAEA has been developing a finite element analysis code for assembled structure to accurately evaluate the structural integrity of nuclear power plant in its entirety under seismic events. Because nuclear power plant is very huge assembled structure with tens of millions of mechanical components, the finite element model of each component is assembled into one structure and non-conforming meshes of mechanical components are bonded together inside the code. The main technique to bond these mechanical components is triple sparse matrix multiplication with multiple point constrains and global stiffness matrix. In our code, this procedure is conducted in a component by component manner, so that the working memory size and computing time for this multiplication are available on the current computing environment. As an illustrative example, seismic simulation of a real nuclear reactor with one thousand mechanical components was conducted.
Yoshida, Hiroyuki; Suzuki, Takayuki*; Takase, Kazuyuki; Koizumi, Yasuo*
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
Misawa, Takeharu; Yoshida, Hiroyuki; Takase, Kazuyuki
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
Suzudo, Tomoaki; Golubov, S.*; Stoller, R.*; Yamaguchi, Masatake; Tsuru, Tomohito; Kaburaki, Hideo
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
Molecular dynamics is a useful tool to analyze cascade damage in structural materials of nuclear devices, but the time scale accessible to molecular dynamics is 100 ps. Kinetic Monte Carlo annealing simulation of cascade damage is useful for analyzing the longer time development of cascade damage. We conducted a series of such annealing simulations in -Fe. The surviving displacement ratio to the NRT displacements before annealing is 0.3 in the case of primary knock-on atom's energy more than 10 keV, but it decreased by 30 % through the annealing at 300 K because of recombination of vacancies and self-interstitial atoms, and the recombination ratio increased as the annealing temperature increased. These results are meaningful when applied to the simulation of accumulation of cascades using rate theory. This work is useful for R&D of nuclear materials.
Sakurai, Takeshi; Kosako, Kazuaki*; Mori, Takamasa
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 8 Pages, 2010/10
Nishizawa, Masato; Suzuki, Takashi; Nagai, Haruyasu; Togawa, Orihiko
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
Suzuki et al. (,
, 268-275, 2008) estimated that more than 80% of Iodine-129 (
I) in seawater in the Japan Sea came from nuclear fuel reprocessing plants. Considering the distance from the main nuclear reprocessing plants in Europe to the Japan Sea and the time scales of atmospheric and ocean circulations, large portion of
I in the Japan Sea is presumed to be transported through the atmosphere. In the present study, a global chemical transport model, MOZART-4, is applied to investigate the behavior of
I emitted from nuclear fuel reprocessing plants in Europe (Sellafield in the UK and La Hague in France) and to estimate the distribution in the atmosphere and deposition in remote sites. The result of numerical simulation for more than fifty-year period from the 1950s is validated by comparison with measurements of
I around the world and analyzed to clarify the characteristic of the distributions of concentration and deposition of
I. The modeled concentrations of
I in precipitation in Europe and depositions in Japanese waters are in the same order as measurements. The emitted
I to the atmosphere is distributed and deposited all over the Northern Hemisphere due to the prevailing westerlies. The emission of
I to the atmosphere is thus important in considering the transport and deposition of
I to remote sites.
Van Rooijen, W. F. G.*; Hazama, Taira; Takeda, Toshikazu*
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 14 Pages, 2010/10
For the reactor physics analysis of fast critical assemblies as well as LMFBRs, the diffusion coefficient is one of the required pieces of data. In the present analysis, the diffusion coefficient is determined using the Benoist-formalism, which is based on directional collision probabilities. For LMFBR analysis including void regions, the Benoist-formalism breaks down if two-dimensional (slab or slab-like) void regions are present. Furthermore, the Benoist-style assumption of zero buckling is questionable in fast reactors. Research is being done to identify improved cell calculations, in order to calculate the diffusion coefficient in one- and two-dimensional unit cells containing real void regions.
Suzuki, Yoshio; Kushida, Noriyuki; Tatekawa, Takayuki; Teshima, Naoya; Caniou, Y.*; Guivarch, R.*; Dayde, M.*; Ramet, P.*
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
The "Research and Development of International Matrix-Solver Prediction System (REDIMPS)" project, which is founded by the Strategic Japanese-French Cooperative Program on "Information and Communications Technology including Computer Science" with CNRS and JST, aims at improving the "Test for Large System of Equations (TLSE)" sparse linear algebra expert system by establishing an international grid computing environment between Japan and France. Here, we have established the interoperable environment between French and Japanese grid middleware (DIET and AEGIS), and have confirmed that TLSE can rely on this French and Japanese interoperable environment for researchers to select a matrix-solver suitable to their each application program. By this study, we proposed to the French and Japanese researchers the environment in which they can obtain useful information for the improvement of their application program.
Kushida, Noriyuki
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
Several applications in nuclear energy field require dedicated supercomputing environment (DSE). One of the greatest common aspects of DES applications is that they have to complete tasks within deadlines. High-end supercomputers are not suitable for DSE applications, because it is publically shared facilities and allotted time for a person is quite limited. Moreover, owning a high end supercomputer is not realistic because of the high price of supercomputers. On the other hand, many core processors have good cost performance and therefore they are suitable for DSE. In the present paper, the author describes the feasibility of many-core processors for such purpose by using two examples: (1) Fusion reactor monitoring and (2) Infrasound propagation analysis.
Nishida, Akemi; Araya, Fumimasa; Kushida, Noriyuki; Kondo, Makoto; Sakai, Michiya*; Shiogama, Yuzo*
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
The objective of this research is to contribute to the seismic design evaluation of nuclear facilities through the construction of a numerical evaluation system which is able to evaluate both global and local behaviors of facilities under severe seismic events. As one of the technology components to realize this objective, we are developing a physical model describing the dynamic interaction characteristics of component connections, called as the elastic-plastic connection model. We focused on the joints of the support structures of the component and the building in nuclear plants which generally designed as fixed/pinned boundaries, and tried to consider their dynamic interaction effects. In this paper, we show the proposal of the elastic-plastic connection model and the application of the model to a numerical simulation using a real plant data. The precision of the model was optimized by adjusting its parameters using the data obtained in the experiment.
Yamaguchi, Masatake; Ebihara, Kenichi; Itakura, Mitsuhiro; Suzudo, Tomoaki; Kaburaki, Hideo
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
It is not known in detail how much solute atoms segregate in grain boundaries of metals and how much the cohesive energy (work of fracture) of grain boundary is decreased by the segregation. From first-principles, we calculated the segregation energy of some solute elements like boron (B), carbon (C), phosphorous (P), and sulfur (S) in bcc Fe Sigma 3 (111) symmetrical tilt grain boundary with varying the segregation density. We find that these elements can segregate up to a high concentration in the grain boundary. We also find that the segregation energy on the fracture surface is significantly larger than that in the grain boundary for the embrittling elements like P and S. On the contrary, the cohesive energy is increased by B and C segregation. The increase-decrease rate in the calculated cohesive energy by solute segregation is found to be well correlated with experimentally observed shift in ductile-to-brittle transition temperature by solute segregation.
Ebihara, Kenichi; Itakura, Mitsuhiro; Yamaguchi, Masatake; Kaburaki, Hideo; Suzudo, Tomoaki
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
The decohesion model in which hydrogen segregating at grain boundaries reduces their strength is considered to explain hydrogen embrittlement of steels. Therefore in order to understand hydrogen embrittlement from this model, stress and hydrogen concentration at grain boundaries need be evaluated under the fracture condition for tensile test specimens. From this consideration, we evaluated the stress and the hydrogen concentration at grain boundaries in the three-dimensional polycrystalline model which was generated by Voronoi tessellation. The different crystallographic orientation was given to each grain. Extracted data from the calculation in the notched round-bar specimen model under the tensile test condition was given to the polycrystalline model as the boundary condition. As a result, it was found that the valuated stress does not reach the fracture stress which was estimated under the condition of the evaluated hydrogen concentration by first principles calculation.
Fukushima, Masahiro; Kitamura, Yasunori; Kugo, Teruhiko; Yamane, Tsuyoshi; Ando, Masaki; Chiba, Go; Ishikawa, Makoto; Okajima, Shigeaki
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 6 Pages, 2010/10
Kurosaki, Ken*; Tanaka, Kosuke; Osaka, Masahiko; Oishi, Yuji*; Muta, Hiroaki*; Uno, Masayoshi*; Yamanaka, Shinsuke*
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
It is important to understand the behavior of fission products and actinides under irradiation. In the present study, the chemical states of fission products and actinides in irradiated oxide fuels were evaluated by both thermodynamic equilibrium calculation and post-irradiation examination.
Nakatsuka, Toru; Misawa, Takeharu; Yoshida, Hiroyuki; Takase, Kazuyuki
Proceedings of Joint International Conference of 7th Supercomputing in Nuclear Application and 3rd Monte Carlo (SNA + MC 2010) (USB Flash Drive), 4 Pages, 2010/10
To evaluate thermal hydraulic characteristics of a tight-lattice fuel bundle of supercritical water reactor (Super Fast Reactor), a simplified 19-rod fuel assembly was analyzed with a three-dimensional two-fluid model analysis code ACE-3D. In this calculation, a one-twelfth model is adopted as the computational domain taking advantage of symmetry. As the boundary conditions, mass velocity, inlet enthalpy and power distribution are to be the same as the steady state condition of the reactor. Cross-sectional local power distribution in the fuel assembly is set to be flat. Effect of grid spacers is taken into account in the analysis. Maximum clad surface temperature (MCST) is observed at the position facing to the narrowest gap on the center rod near the outlet and the value is 628C that is almost the same as results without grid spacers. The predicted MCST satisfies a thermal design criteria to ensure fuel and cladding integrity: the MCST should be less than 650
C.