A5101413965- Computational Fluid Dynamics and Aerodynamics
- Gas Dynamics and Kinetic Theory
- Combustion and flame dynamics
- Fluid Dynamics and Turbulent Flows
- Turbomachinery Performance and Optimization
- Heat transfer and supercritical fluids
- nanoparticles nucleation surface interactions
- Phase Equilibria and Thermodynamics
- Parallel Computing and Optimization Techniques
- Subcritical and Supercritical Water Processes
- Plasma and Flow Control in Aerodynamics
- Lattice Boltzmann Simulation Studies
- Embedded Systems Design Techniques
- Numerical Methods and Algorithms
- Particle Dynamics in Fluid Flows
- Advanced Data Storage Technologies
- Spacecraft and Cryogenic Technologies
- Aerodynamics and Fluid Dynamics Research
- Advanced Numerical Methods in Computational Mathematics
- Cyclone Separators and Fluid Dynamics
- Refrigeration and Air Conditioning Technologies
- Icing and De-icing Technologies
- Fluid Dynamics Simulations and Interactions
- Nuclear Engineering Thermal-Hydraulics
- Fluid Dynamics and Vibration Analysis
Tohoku University
2014-2023
Mitsubishi Motors (Japan)
2011-2013
Tottori University
2013
The University of Tokyo
2012
Hokkaido University
1997-2011
Teikyo University
2010-2011
Toshiba (Japan)
2011
Tohoku Gakuin University
2008
Tohoku University Hospital
2007
Vaughn College of Aeronautics and Technology
2004
Rat tibia tissue into which calcium ion (Ca2+)-implanted titanium was surgically placed histologically analyzed to investigate the performance of Ca2+-implanted as a biomaterial. Calcium ions were implanted only one side plates at 1017 ions/cm2 and Ca2+-treated rat for 2, 8, 18 days. Tetracycline calcein used hard-tissue labels. After excision tibia, tissues fixed, stained, embedded in polymethyl methacrylate, sliced. The specimens observed using fluorescence microscope. A larger amount new...
Stencil computation is one of the important kernels in scientific computations. However, sustained performance limited owing to restriction on memory bandwidth, especially multicore microprocessors and graphics processing units (GPUs) because their small operational intensity. In this paper, we present a custom computing machine (CCM), called scalable streaming-array (SSA), for high-performance stencil computations with multiple field-programmable gate arrays (FPGAs). We design SSA based...
The purpose of the “International Wet Steam Modeling Project” is to review ability computational methods predict condensing steam flows. results numerous wet-steam are compared with each other and experimental data for several nozzle test cases. spread computed quite noticeable present paper endeavours explain some reasons this. Generally, however, confirm that reasonable agreement experiment obtained by using classical homogeneous nucleation theory corrected non-isothermal effects, combined...
The tissue response to hat-stacked carbon nanofibers (H-CNFs) was evaluated. H-CNFs were implanted in the subcutaneous of rats. Histological and ultrastructural investigations carried out by transmission electron microscopy. Although many macrophages foreign body giant cells seen around H-CNFs, no severe inflammatory such as necrosis observed. Some observed lysosomal vacuoles phagocytes. These results showed that not strong prophlogistic substances englobed vivo.
This paper presents an FPGA-based flow solver based on the systolic architecture. We show that fractional-step method employing central difference schemes can be expressed as a algorithm, and therefore architecture is suitable for dedicated processor to solver. have designed 2D array of cells, each which has micro-programmable data-path containing MAC (multiplication accumulation) unit local memory store necessary data computational fluid dynamics. With ALTERA Stratix II FPGA, we implemented...
High-performance and low-power computation is required for large-scale fluid dynamics simulation. Due to the inefficient architecture structure of CPUs GPUs, they now have a difficulty in improving power efficiency target application. Although FPGAs become promising alternatives power-efficient high-performance due their new having floating-point (FP) DSP blocks, relatively narrow memory bandwidth requires an appropriate way fully exploit advantage. This paper presents design scalable...
This paper presents a performance model of an LBM accelerator to be implemented on tightly-coupled FPGA cluster. In strong scaling, each node has smaller computation as the nodes increase, and consequently communication overhead becomes apparent limits scalability. Our cluster 1D ring accelerator-domain network (ADN) which allows FPGAs send receive data with low overhead. We propose architecture its stream appropriate use ADN. formulate sustained-performance accelerator, consists three cases...
Stencil computation is one of the important kernels in scientific computations, however, sustained performance limited by memory bandwidth especially on multi-core microprocessors and GPGPUs due to its small operationalintensity. In this paper, we propose a scalable streaming-array (SSA) simple soft-processors for high-performance stencil multiple FPGAs. The SSA architecture allows multi-device system have linear scalability computing deeply pipelining with constant an external-memory. We...
For scientific numerical simulation that requires a relatively high ratio of data access to computation, the scalability memory bandwidth is key performance improvement, and therefore custom-computing machines (CCMs) are one promising approaches provide bandwidth-aware structures tailored for individual applications. In this article, we propose scalable FPGA-array with bandwidth-reduction mechanism (BRM) implement high-performance power-efficient CCMs simulations based on finite difference...
A preconditioned numerical method for gas-liquid two-phase flows is applied to solve cavitating flow. The present employs a finite-difference of the dual time-stepping integration procedure and Roe’s flux difference splitting approximation with MUSCL-TVD scheme. homogeneous equilibrium cavitation model used. density-based permits simple treatment whole flow field, including wave propagation, large density changes incompressible characteristics at low Mach number. Two-dimensional internal...
This work is conducted towards the accurate prediction of compressor tip vortex. Accurate computation highly vortical flow affected by many parameters, such as numerical scheme and turbulence model. In this effect studied using mesh refinement study comparison results from central a newly developed high order upwind scheme. Behaviors methods in vortex region are also theoretically numerically analyzed. It found computed significantly resolution dissipation. Currently widely strategy, i.e.,...
Although computational performance is often limited by insufficient bandwidth to/from an external memory, it not easy to physically increase off-chip memory bandwidth. In this study, we propose a hardware-based compression technique that can be applied field-programmable gate array-- (FPGA) based high-performance computation with logically wider effective Our proposed hardware approach boost the of FPGA-based stream computations applying data effectively transfer more streams. To apply via...
The paper presents mathematical modelling of steam condensation in the atmospheric air transonic flows. A single-fluid model was employed. validated against in-house experimental studies internal flow nozzles and benchmark tests available literature i.e., RAE2822 airfoil. impact relative humidity on process under conditions examined. shows that is non-negligible has to be taken consideration simulations. presence contaminants form small particles considered too. Their crucial role change...