• IEMEK Journal of Embedded Systems and Applications
• /
• v.9 no.6
• /
• pp.335-344
• /
• 2014

As a kind of 2D spatial coordinate transform, image warping is a basic image processing technique utilized in various applications. Though image warping algorithm is composed of relatively simple operations such as memory accesses and computations of weighted average, real-time implementations on embedded vision systems suffer from limited computational power because the simple operations are iterated as many times as the number of pixels. This paper presents a real-time implementation of a look-up table(LUT)-based image warping using an FPGA. In order to ensure sufficient data transfer rate from memories storing mapping LUT and image data, appropriate memory devices are selected by analyzing memory access patterns in an LUT-based image warping using backward mapping. In addition, hardware structure of a parallel and pipelined architecture is proposed for fast computation of bilinear interpolation using fixed-point operations. Accuracy of the implemented hardware is verified using a synthesized test image, and an application to real-time lens distortion correction is exemplified.

• Transactions of the Korean Society of Mechanical Engineers B
• /
• v.23 no.8
• /
• pp.1022-1030
• /
• 1999

A numerical study was carried out to analyze the effect of flow distribution of stirred part and plug flow part on combustion efficiency at the coal gasification process in an entrained bed coal reactor. The model of computation was based on gas phase eulerian balance equations of mass and momentum. The solid phase was described by lagrangian equations of motion. The $k-{\varepsilon}$ model was used to calculate the turbulence flow and eddy dissipation model was used to describe the gas phase reaction rate. The radiation was solved using a Monte-Carlo method. One-step parallel two reaction model was employed for the devolatilization process of a high volatile bituminous Kideco coal. The computations agreed well with the experiments, but the flame front was closer to the burner than the measured one. The flow distribution of a stirred part and a plug flow part in a reactor was a function of the magnitude of recirculation zone resulted from the swirl. The combustion efficiency was enhanced with decreasing stirred part and the maximum value was found around S=1.2, having the minimum stirred part. The combustion efficiency resulted from not only the flow distribution but also the particle residence time through the hot reaction zone of the stirred part, in particular for the weak swirl without IRZ(internal recirculation zone) and the long lifted flame.

• Transactions of the Korean Society for Noise and Vibration Engineering
• /
• v.19 no.5
• /
• pp.482-490
• /
• 2009

In this paper, the author proposes a new method for acoustic radiation optimum design to minimize noise from a vibrating panel-like structure using a collaborative population-based search method called the particle swarm optimization algorithm(PSOA). The PSOA is a parallel evolutionary computation technique initially developed by Kennedy and Eberhart. The acoustic radiation optimization method based on the PSOA consists of two processes. In the first process, the acoustic radiation analysis by an integrated p-version FEM/BEM, which was developed by using MATLAB, is performed to evaluate the exterior acoustic radiation field of the panel. The second process is to search the optimum design variables: 1) Shape of Bezier curves and 2) Shape and position of ribs, to minimize noise from the panel using the PSOA. The optimization method based on the PSOA is compared to that based on the steady state genetic algorithm(SSGA) in order to verify the effectiveness and validity of the optimal solution by PSOA. Finally, it is shown that the optimal designs of the panel obtained by using the PSOA can achieve effective reductions in radiated sound power.

• The KIPS Transactions:PartB
• /
• v.11B no.6
• /
• pp.735-742
• /
• 2004

We present a new use of common graphics hardware to perform a faster artificial neural network. And we examine the use of GPU enhances the time performance of the image processing system using neural network, In the case of parallel computation of multiple input sets, the vector-matrix products become matrix-matrix multiplications. As a result, we can fully utilize the parallelism of GPU. Sigmoid operation and bias term addition are also implemented using pixel shader on GPU. Our preliminary result shows a performance enhancement of about thirty times faster using ATI RADEON 9800 XT board.

• The Transactions of The Korean Institute of Electrical Engineers
• /
• v.67 no.2
• /
• pp.270-276
• /
• 2018

Apache Spark is one of the high performance in-memory computing frameworks for big-data processing. Recently, to improve the performance, general-purpose computing on graphics processing unit(GPGPU) is adapted to Apache Spark framework. Previous Spark-GPGPU frameworks focus on overcoming the difficulty of an implementation resulting from the difference between the computation environment of GPGPU and Spark framework. In this paper, we propose a Spark framework based on a heterogenous pipeline computing with OpenCL to further improve the performance. The proposed framework overlaps the Java-to-Native memory copies of CPU with CPU-GPU communications(DMA) and GPU kernel computations to hide the CPU idle time. Also, CPU-GPU communication buffers are implemented with switching dual buffers, which reduce the mapped memory region resulting in decreasing memory mapping overhead. Experimental results showed that the proposed Spark framework based on a heterogenous pipeline computing with OpenCL had up to 2.13 times faster than the previous Spark framework using OpenCL.

• 한국전산유체공학회:학술대회논문집
• /
• 2005.10a
• /
• pp.163-167
• /
• 2005

A research to evaluate efficiency of design optimization was performed for aerodynamic design optimization problem in distributed computing environment. The aerodynamic analyses which take most of computational work during design optimization were divided into several jobs and allocated to associated PC clients through network. This is not a parallel process based on domain decomposition rather than a simultaneous distributed-analyses process using network-distributed computers. GBOM(gradient-based optimization method), SAO(Sequential Approximate Optimization) and RSM(Response Surface Method) were implemented to perform design optimization of transonic airfoil and to evaluate their efficiencies. One dimensional minimization followed by direction search involved in the GBOM was found an obstacle against improving efficiency of the design process in distributed computing environment. The SAO was found quite suitable for the distributed computing environment even it has a handicap of local search. The RSM is apparently the fittest for distributed computing environment, but additional trial and error works needed to enhance the reliability of the approximation model are annoying and time-consuming so that they often impair the automatic capability of design optimization and also deteriorate efficiency from the practical point of view.

• Advances in Energy Research
• /
• v.7 no.1
• /
• pp.35-52
• /
• 2020

The paper proposes a hybrid approach of artificial bee colony (ABC) and grey wolf optimizer (GWO) algorithm for multi-objective and multidimensional engine optimization of a converted plug-in hybrid electric vehicle. The proposed strategy is used to optimize all emissions along with brake specific fuel consumption (FC) for converted parallel operated diesel plug-in hybrid electric vehicle (PHEV). All emissions particulate matter (PM), nitrogen oxide (NOx), carbon monoxide (CO) and hydrocarbon (HC) are considered as optimization parameters with weighted factors. 70 hp engine data of NOx, PM, HC, CO and FC obtained from Oak Ridge National Laboratory is used for the study. The algorithm is initialized with feasible solutions followed by the employee bee phase of artificial bee colony algorithm to provide exploitation. Onlooker and scout bee phase is replaced by GWO algorithm to provide exploration. MATLAB program is used for simulation. Hybrid ABC-GWO algorithm developed is tested extensively for various values of speeds and torque. The optimization performance and its environmental impact are discussed in detail. The optimization results obtained are verified by real data engine maps. It is also compared with modified ABC and GWO algorithm for checking the effectiveness of proposed algorithm. Hybrid ABC-GWO offers combine benefits of ABC and GWO by reducing computational load and complexity with less computation time providing a balance of exploitation and exploration and passes repeatability towards use for real-time optimization.

• Proceedings of the Korea Contents Association Conference
• /
• 2003.05a
• /
• pp.461-465
• /
• 2003

Grid-computing on networked computers is increasingly applied to the variety of large-scale computation problem. Several software systems are developed for providing the application programmers with computers available. However, these systems are not web-based systems, lacks a collaborative environment or do not supply the real-time visualization facility. Web technology is become the general technology on the development of network application, in particular, because the interface can be made platform independent. In this paper, we propose the web-based framework for executing the parallel SPMD application written MPI. Also, a web-based collaborative environment is development with a real-time visualization technology.

• Proceedings of the Korean Society of Propulsion Engineers Conference
• /
• 2009.05a
• /
• pp.289-292
• /
• 2009

Injector dynamics of multi swirl injectors in an annular combustor have been investigated by LES(Large Eddy Simulation) turbulent model with MPI parallel computation technique. The present study employs the LM6000 lean premixed swirl-stabilized annular combustor. Real shape combustor is simulated in order to investigate the detail interaction mechanism among multi-injectors. The strong vortex breakdown occurs at the impinging surface between the adjacent injectors so that the complex and strong oscillatory pressure propagates inside of the combustor. Tangential pressure fluctuation mode was captured by including multi injectors in computational domain.

• Journal of The Korean Astronomical Society
• /
• v.47 no.3
• /
• pp.87-98
• /
• 2014

We develop a parallel cosmological hydrodynamic simulation code designed for the study of formation and evolution of cosmological structures. The gravitational force is calculated using the TreePM method and the hydrodynamics is implemented based on the smoothed particle hydrodynamics. The initial displacement and velocity of simulation particles are calculated according to second-order Lagrangian perturbation theory using the power spectra of dark matter and baryonic matter. The initial background temperature is given by Recfast and the temperature uctuations at the initial particle position are assigned according to the adiabatic model. We use a time-limiter scheme over the individual time steps to capture shock-fronts and to ease the time-step tension between the shock and preshock particles. We also include the astrophysical gas processes of radiative heating/cooling, star formation, metal enrichment, and supernova feedback. We test the code in several standard cases such as one-dimensional Riemann problems, Kelvin-Helmholtz, and Sedov blast wave instability. Star formation on the galactic disk is investigated to check whether the Schmidt-Kennicutt relation is properly recovered. We also study global star formation history at different simulation resolutions and compare them with observations.