Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Initialising ...
Doda, Norihiro; Nakamine, Yoshiaki*; Kuwagaki, Kazuki; Hamase, Erina; Kikuchi, Norihiro; Yoshimura, Kazuo; Matsushita, Kentaro; Tanaka, Masaaki
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 28, 5 Pages, 2023/05
As a part of the development of the "Advanced Reactor Knowledge- and AI-aided Design Integration Approach through the whole plant lifecycle (ARKADIA)" to automatically optimize the life cycle of innovative nuclear reactors including fast reactors, ARKADIA-design is being developed to support the optimization of fast reactor in the conceptual design stage. ARKADIA-Design consists of three systems (Virtual plant Life System (VLS), Evaluation assistance and Application System (EAS), and Knowledge Management System (KMS)). A design optimization framework controls the connection between the three systems through the interfaces in each system. This paper reports on the development of an optimization analysis control function that performs design optimization analysis combining plant behavior analysis by VLS and optimization study by EAS.
Onodera, Naoyuki; Idomura, Yasuhiro; Hasegawa, Yuta; Asahi, Yuichi; Inagaki, Atsushi*; Shimose, Kenichi*; Hirano, Kohin*
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 28, 4 Pages, 2023/05
We have developed a multi-scale wind simulation code named CityLBM that can resolve entire cities to detailed streets. CityLBM enables a real time ensemble simulation for several km square area by applying the locally mesh-refined lattice Boltzmann method on GPU supercomputers. On the other hand, real-world wind simulations contain complex boundary conditions that cannot be modeled, so data assimilation techniques are needed to reflect observed data in the simulation. This study proposes an optimization method for ground surface temperature bias based on an ensemble Kalman filter to reproduce wind conditions within urban city blocks. As a verification of CityLBM, an Observing System Simulation Experiment (OSSE) is conducted for the central Tokyo area to estimate boundary conditions from observed near-surface temperature values.
Hasegawa, Yuta; Onodera, Naoyuki; Asahi, Yuichi; Idomura, Yasuhiro
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 27, 4 Pages, 2022/06
We developed GPU implementation of ensemble data assimilation (DA) using the local ensemble transform Kalman filter (LETKF) with the lattice Boltzmann method (LBM). The performance test was carried out upto 32 ensembles of two-dimensional isotropic turbulence simulations using the D2Q9 LBM. The computational cost of the LETKF was less than or nearly equal to that of the LBM upto eight ensembles, while the former exceeded the latter at larger ensembles. At 32 ensembles, their computational costs per cycle were respectively 28.3 msec and 5.39 msec. These results suggested that further speedup of the LETKF is needed for practical 3D LBM simulations.
Sugihara, Kenta; Onodera, Naoyuki; Idomura, Yasuhiro; Yamashita, Susumu
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 27, 5 Pages, 2022/06
The phase-field method has been successfully applied to various multi-phase flow problems as an interface tracking method for gas-liquid interfaces. However, the accuracy of the phase-field method depends on hyper-parameters, which are empirically adjusted for each problem. The phase-field method sustains sharp interfaces by the balance between the numerical viscosity of the advection term and the interface modification by the diffusion and anti-diffusion terms. Based on this fact, we propose a method for deriving the optimal hyper-parameters in a non-empirical manner by performing a basic error analysis of the interface advection.
Onodera, Naoyuki; Idomura, Yasuhiro; Hasegawa, Yuta; Shimokawabe, Takashi*; Aoki, Takayuki*
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 27, 4 Pages, 2022/06
We have developed a wind simulation code named CityLBM to realize wind digital twins. Mesoscale wind conditions are given as boundary conditions in CityLBM by using a nudging data assimilation method. It is found that conventional approaches with constant nudging coefficients fail to reproduce turbulent intensity in long time simulations, where atmospheric stability conditions change significantly. We propose a dynamic parameter optimization method for the nudging coefficient based on a particle filter. CityLBM was validated against plume dispersion experiments in the complex urban environment of Oklahoma City. The nudging coefficient was updated to reduce the error of the turbulent intensity between the simulation and the observation, and the atmospheric boundary layer was reproduced throughout the day.
Asahi, Yuichi; Onodera, Naoyuki; Hasegawa, Yuta; Shimokawabe, Takashi*; Shiba, Hayato*; Idomura, Yasuhiro
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 27, 5 Pages, 2022/06
We have ported the GPU accelerated Lattice Boltzmann Method code "CityLBM" to AMD MI100 GPU. We present the performance of CityLBM achieved on NVIDIA P100, V100, A100 GPUs and AMDMI100 GPU. Using the host to host MPI communications, the performance on MI100 GPU is around 20% better than on V100 GPU. It has turned out that most of the kernels are successfully accelerated except for interpolation kernels for Adaptive Mesh Refinement (AMR) method.
Doda, Norihiro; Nakamine, Yoshiaki*; Igawa, Kenichi*; Iwasaki, Takashi*; Murakami, Satoshi*; Tanaka, Masaaki
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 27, 6 Pages, 2022/06
As a part of the development of the "Advanced Reactor Knowledge- and AI-aided Design Integration Approach through the whole plant lifecycle (ARKADIA)" to automatically optimize the life cycle of innovative nuclear reactors including fast reactors, ARKADIA-design is being developed to support the optimization of fast reactor design in the conceptual stage. ARKADIA-Design consists of three systems (Virtual plant Life System (VLS), Evaluation assistance and Application System (EAS), and Knowledge Management System (KMS)). A design optimization framework controls the cooperation between the three systems through the interfaces in each system. This paper reports on the development status of the "VLS interface," which has a control function of coupling analysis codes in VLS.
Hasegawa, Yuta; Aoki, Takayuki*; Kobayashi, Hiromichi*; Idomura, Yasuhiro; Onodera, Naoyuki
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 26, 6 Pages, 2021/05
We introduce an improved domain partitioning method called "tree cutting approach" for the aerodynamics simulation code based on the lattice Boltzmann method (LBM) with the forest-of-octrees-based local mesh refinement (LMR). The conventional domain partitioning algorithm based on the space-filling curve (SFC), which is widely used in LMR, caused a costly halo data communication which became a bottleneck of our aerodynamics simulation on the GPU-based supercomputers. Our tree cutting approach adopts a hybrid domain partitioning with the coarse structured block decomposition and the SFC partitioning in each block. This hybrid approach improved the locality and the topology of the partitioned sub-domains and reduced the amount of the halo communication to one-third of the original SFC approach. The code achieved 
speedup on 8 GPUs, and achieved 
speedup at the performance of 2207 MLUPS (mega-lattice update per second) on 128 GPUs with strong scaling test.
Matsushita, Kentaro; Fujisaki, Tatsuya*; Ezure, Toshiki; Tanaka, Masaaki; Uchida, Mao*; Sakai, Takaaki*
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 26, 6 Pages, 2021/05
For the gas entrainment vortex at the free surface in sodium-cooled fast reactors, development of the numerical analysis method to evaluate amount of the gas entrainment from the free surface has been developing. In this paper, the automatic creation of analysis meshes which can suppress the calculation cost while maintaining the prediction accuracy of the vortex shape is investigated, and the adaptive mesh refinement (AMR) method is examined to the creation of analysis mesh applying to the unsteady vortex system. The refined mesh based on the criterion evaluated by vorticity, Q-value as second invariant of the velocity and the discriminant for the eigen equation of the velocity gradient tensor is considered, and it found that the AMR method based on Q-value can refine the analysis meshes most efficiently.
Doda, Norihiro; Uwaba, Tomoyuki; Nemoto, Toshiyuki*; Yokoyama, Kenji; Tanaka, Masaaki
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 26, 4 Pages, 2021/05
For design optimization of fast reactors, in order to consider the feedback reactivity due to thermal deformation of the core when the core temperature rises, which could not be considered in the conventional design analysis, a neutronics, thermal-hydraulics, and structure mechanics coupled analysis method has been developed. Neutronics code, plant dynamics code, and structural mechanics code are coupled by a control module in python script. This paper outlines the coupling method of analysis codes and the results of its application to an experiment in an actual plant.
Onodera, Naoyuki; Idomura, Yasuhiro; Hasegawa, Yuta; Shimokawabe, Takashi*; Aoki, Takayuki*
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 26, 3 Pages, 2021/05
We develop a mixed-precision preconditioner for the pressure Poisson equation in a two-phase flow CFD code JUPITER-AMR. The multi-grid (MG) preconditioner is constructed based on the geometric MG method with a three- stage V-cycle, and a cache-reuse SOR (CR-SOR) method at each stage. The numerical experiments are conducted for two-phase flows in a fuel bundle of a nuclear reactor. The MG-CG solver in single-precision shows the same convergence histories as double-precision, which is about 75% of the computational time in double-precision. In the strong scaling test, the MG-CG solver in single-precision is accelerated by 1.88 times between 32 and 96 GPUs.
Asahi, Yuichi; Hatayama, Sora*; Shimokawabe, Takashi*; Onodera, Naoyuki; Hasegawa, Yuta; Idomura, Yasuhiro
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 26, 4 Pages, 2021/05
We develop a convolutional neural network model to predict the multi-resolution steady flow. Based on the state-of-the-art image-to-image translation model Pix2PixHD, our model can predict the high resolution flow field from the signed distance function. By patching the high resolution data, the memory requirements in our model is suppressed compared to Pix2PixHD.
Hasegawa, Yuta; Onodera, Naoyuki; Idomura, Yasuhiro
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 25, 4 Pages, 2020/06
We developed a GPU-based CFD code using a mesh-refined lattice Boltzmann method (LBM), which enables ensemble simulations for wind and plume dispersion in urban cities. The code is tuned for Pascal or Volta GPU architectures, and is able to perform real-time wind simulations with several kilometers square region and several meters of grid resolution. We examined the developed code against the field experiment JU2003 in Oklahoma City. In the comparison, wind conditions showed good agreements, and the ensemble-averaged and maximum values of tracer concentration satisfied the factor 2 agreements.
Onodera, Naoyuki; Idomura, Yasuhiro; Ali, Y.*; Shimokawabe, Takashi*; Aoki, Takayuki*
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 25, 4 Pages, 2020/06
We have developed the stencil-based CFD code JUPITER for simulating three-dimensional multiphase flows. A GPU-accelerated Poisson solver based on the preconditioned conjugate gradient (P-CG) method with a multigrid preconditioner was developed for the JUPITER with block-structured AMR mesh. All Poisson kernels were implemented using CUDA, and the GPU kernel function is well tuned to achieve high performance on GPU supercomputers. The developed multigrid solver shows good convergence of about 1/7 compared with the original P-CG method, and 
3 speed up is achieved with strong scaling test from 8 to 216 GPUs on TSUBAME 3.0.
Doda, Norihiro; Hamase, Erina; Yokoyama, Kenji; Tanaka, Masaaki
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 25, 4 Pages, 2020/06
With the aim of advancing the design optimization in fast reactors, neutronics and thermal-hydraulics coupled analysis method which can consider the temporal change of neutron flux distribution in the core has been developed. A three-dimensional neutronics analysis code and a plant dynamics analysis code are coupled on a platform using Python programing. In this report, outlines of the coupling method of analysis codes, the results of its application to the actual plant under a virtual accidental condition, and the future development is described.
Tanaka, Masaaki
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 22, 4 Pages, 2017/05
In the development of the simulation code and the numerical estimation for high cycle thermal fatigue on a structure caused by thermal striping phenomena in sodium cooled fast reactors, implementation of verification and validation (V&V) process is indispensable. A procedure named V2UP (Verification and Validation plus Uncertainty quantification and Prediction) has been made by referring to the existing guidelines regarding the V&V and the methodologies of the safety assessment. In this paper, a challenging installation of quality management procedures into the V2UP procedure is attempted based on the JSCES Standard for "A Model Procedure for Engineering Simulation".
Suzuki, Yoshio; Nishida, Akemi; Uzawa, Ken; Nakajima, Norihiro
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 17, 4 Pages, 2012/05
Our goal is to evaluate the validity of results obtained from seismic simulation of entire nuclear plant, and then to estimate real behavior. In this study, we have established a trial tool to enable a zooming analysis from a combining analysis of ground-building-equipments (beam model) to an assembly structural analysis of equipments (solid model) in order to compare results of beam model and solid model. By applying the tool to analysis of a cantilever model with various cross sections, we have confirmed that results of solid model well coincide with theoretical solutions and results of beam model.
Suzuki, Yoshio; Miyamura, Hiroko
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 16, 4 Pages, 2011/05
Our goal is to evaluate the validity of results obtained from seismic simulation of entire nuclear power plant, and then to estimate real phenomena. In this study, we have established a trial system to evaluate the validity by automatically abstracting a characteristic spatial profile of physical values in consideration of uncertainty caused by discretization, as one of methods to reveal factors of various uncertainties included in the simulation. By applying the system to a seismic simulation, we have confirmed that various characteristic spatial profiles of stress with error estimation are automatically classified and abstracted from the simulation result.
Yamada, Tomonori
Keisan Kogaku Koenkai Rombunshu (CD-ROM), 16, 2 Pages, 2011/05
The Japanese Ministry of Education, Culture, Sports, Science and Technology (MEXT) has been developing the next-generation supercomputer, which has capability of ten peta floating point operations per second, and will install it at Kobe city. As an application of macro-scale solid mechanics, Japan Atomic Energy Agency and his colleagues proposed a research activity on K entitled "next generation seismic analysis of large plants in its entirety" in Monodukuri field. In this paper, a preliminary study on the finite element analysis for solid mechanics on K is described.
Suzuki, Yoshio; Tatekawa, Takayuki; Kim, G.; Kino, Chiaki; Miyamura, Hiroko; Teshima, Naoya; Hayashi, Sachiko*; Aoyagi, Tetsuo; Takemiya, Hiroshi; Nakajima, Norihiro
Keisan Kogaku Koenkai Rombunshu, 15(2), p.1051 - 1054, 2010/05
We have developed the Atomic Energy Grid InfraStructure (AEGIS) for establishing computational infrastructure for nuclear field. In this R&D we have inherited the knowledge and skills in ITBL Infrastructure developed in the national project ITBL (Information Technology Based Laboratory), which aims at establishment of virtual research environment where supercomputers and data bases are connected by network. Here we have focused on two issues: improvements of safety and usability. For the safety, we have made the authentication mechanism double with both the personal certification and the machine certification. For the usability, we have developed the grid-enabled client API to use grid functions on a user terminal. By those R&Ds, we have successfully contributed to various nuclear researches, such as "full scale 3D vibration simulator for an entire nuclear power plant", "simulation for predicting quake-proof capability of nuclear power plants", and so on.