A CFD-based design method for transonic axial compressor blades was developed based on three-dimensional Navier-Stokes flow physics. The method employs a sectional three-dimensional (S3D) analysis concept where the three-dimensional flow analysis is performed on the grid plane of a span station with spanwise flux components held fixed. The S3D analysis produced flow solutions nearly identical to those of three-dimensional analysis, regardless of the initialization of the flow field. The sectional design based on the S3D analysis can include three-dimensional effects of compressor flows and thus overcome the deficiencies associated with the use of quasi-three-dimensional flow physics in conventional sectional design. The S3D design was first used in the inverse triode to find the geometry that produces a specified target pressure distribution. The method was also applied to optimize the adiabatic efficiency of the blade sections of Rotor 37. A new blade was constructed with the optimized sectional geometries at several span stations and its aerodynamic performance was evaluated with three-dimensional analyses.

Astrophysics has been one of the disciplines which utilize actively supercomputers for their researches. In this paper, studies of astrophysical flows by large scale numerical simulations in supercomputers are discussed. The followings are described: 1) characteristics of astrophysical flows, 2) numerical codes to study astrophysical flows, 3) performance of the codes in parallel supercomputers. A couple of examples of such studies are briefly presented: 1) three-dimensional evolution of the nonlinear Kelvin-Helmholtz instability in a magnetized medium, 2) three-dimensional simulations of astrophysical jets in a magnetized medium.

Two-dimensional steady flowfields generated by slot injection into supersonic flow are numerically simulated by the integration of Navier-stokes equation with two-equation $\kappa-\epsilon$ turbulence model. High-order upwind scheme is used on unstructured adaptive meshes. The numerical results are compared with experimental data in terms of surface static pressure distributions, the length of the upstream separation region, and the height of the Mach surface for steady flowfields with a Mach number of 3.71 and a unit Reynolds number of $5.83\times10^6/m$.

For rapid and abrupt control of a missile in supersonic flight, side jet on a missile body is found to be a useful devise as evidenced by recent missile development at several nations. The magnitude of the side jet and the duration of it decide the level of control of such a missile system. In this paper, the aerodynamic characteristics of the side jet devise itself are examined in terms of key parameters such as the side jet nozzle contour, the chamber pressure and temperature. Specifically attention is focused on the effect of the chamber shape between the straight nozzle and the bended nozzle by 90 degrees. The flow properties as well as the thrust level are compared between the two shapes. Particular attention is paid at the way the nozzle is bended at the joint. Effects of the length and the divergence angle of the nozzle on the thrust magnitude are also quantified among the three different side jet nozzles.

A Numerical simulation for the propulsion of axisymmetric body by contractive and dilative motion is carried out. The present analysis shows that a propulsive force can be obtained in highly viscous fluid by a contractive and dilative motion of axisymmetric body. An axisymmetric analysis code is developed with unstructured grid system for the simulation of complicated motion and geometry. The developed code is validated by comparing with the results of stokes approximation with the problem of uniform flow past a sphere in low Reynolds number($R_n=1$). The validated code is applied to the simulation of contractive and dilative motion of body. The simulation is extended to the analysis of waving surface with projecting part for finding out the difference of hydrodynamic performance according to the variation of waving surface configuration. The present study will be the basic research for the development of the propulsor of an axisymmetric micro-hydro-machine.

An efficient Gauss-Seidel time integration scheme is developed for solving the Euler and Navier-Stokes equations on unstructured meshes. Roe's FDS is used for the explicit residual computations and van Leer's FVS for evaluating implicit flux Jacobian. To reduce the memory requirement to a minimum level, off-diagonal flux Jacobian contributions are repeatedly calculated during the Gauss-Seidel sub-iteration process. Computational results based on the present scheme show that approximately $15\%$ of CPU time reduction is achieved while maintaining the memory requirement level to $50-60\%$ of the original Gauss-Seidel scheme.

An approach of composing a performance optimized computational code is suggested for latest microprocessors. The approach named as localization is a concept of minimizing the access to system's main memory and maximizing the utilization of second level cache that is common to all the latest computer system. The localized compositions of LU-SGS scheme for fluid dynamics were made in three different levels and tested on three different microprocessor architectures most widely used in these days. The test results of localization concept showed a remarkable performance, that is the showing gain up to 4.5 times faster solution than the baseline algorithm $450\%$ for producing an exactly the same solution.

An evaluation of one zero-equation and two one-equation eddy viscosity-transport turbulence closure models as implemented CFDS(Characteristic Flux Difference Splitting ) code is presented herein. Comparisons of Baldwin-Lomax model as zero-equation and Baldwin-Barth and Spalart-Allmaras model as one-equation are presented for three test cases, first inlvolving the 3 dimensional supersonic flow at M=1.98 over tangent ogive cylinder, second involving the 2 dimensional transonic flow at M=0.79 over RAE 2822 airfoil, third involving the 3 dimensional transonic flow at M=0.84 over ONERA M6 wing. The numerical results of CFDS code will also examined through direct comparison with experimental data.

The projectile afterbodies for zero-lift drag reduction has been analyzed using the Navier-Stokes equations with the $\kappa-\epsilon$ turbulence model. The numerical method of a second order upwind scheme has been used on unstructured adaptive meshes. Base drag reduction methods that have been found effective on axisymmetric bodies include boattailing, base bleed, base comustion, locked vortex afterbodies and multistep afterbodies. In this paper, the charateristics of turbulence flow have been studied for geomeries of multistep afterbodies. The important geometrical and flow parameters relevant to the design of such afterbodies have been identified by number, length and height of step. The flow over multistep afterbodies has been analyzed including expansion waves, recompression waves, recirculating flow, shear flow and wake flow. The numerical results have been compared and analyzed with the experimental datum.

본 논문은 삼동선의 선형특성에 대한 간략한 소개와 삼동선의 주선체 좌우에 설치되는 보조 선체의 위치가 조파저항성능에 미치는 영향과 그 최적위치선정을 위한 연구 결과를 보인다. 조파저항 계산기법을 이용하여 보조선체의 길이방향과 횡방향의 위치 변화에 대한 조파저항을 계산하였으며, 그 유용성을 검증하기 위하여 모형시험결과와 함께 비교하였다 본 연구를 통하여 보조선체의 횡방향 위치가 삼동선의 조파저항성능에 미치는 영향은 작으나, 길이방향 위치는 그 성능에 커다란 영향을 미치는 것으로 나타났다. 그리고 주선체의 선수부에서 발생된 커다란 파도의 파저에 보조선체의 선수가 놓이는 것이 저항성능에 가장 유리하며, 선속이 빨라짐에 따라서 그 최적위치는 선미쪽으로 옮겨가는 경향을 보였다.

The computational efficiency of an shape optimization procedure is highly dependent upon the proper selection of shape representation methods and design variables. In this study, shape functions, Bezier and NURBS(non-uniform rational B-splines) curves are selected as configuration generation methods and their efficiencies on the nose shape design of high-speed air vehicles, are compared. The effects of the number of control points, weighting factors and the optimization methods when utilizing the NURBS curves, are investigated. By implementing Bezier and NURBS curves, shapes having lower drag than the optimization case utilizing the shape functions, were obtained, hence it was demonstrated that these curves have better capability in representing the configuration. Efforts will be given to improve the convergence behavior when utilizing the NURBS, hence to reduce the number of Navier-Stokes analysis calculations.

A dual-time stepping algorithm combined with a parallelized multigrid DADI method is presented to predict the dynamic damping coefficients. The Basic Finner model is chosen to validate the prediction capability of the present unsteady Euler method. The linearity of the pitch- and roll-damping coefficients is shown in the low angular rates and the interesting large drop and stiff increment in transonic region for roll-damping coefficients are explained in detail. Through the analysis for the pressure distributions at Mach number 1.0 to 1.2, the sudden drop results from the normal shock and the stiff increment of roll-damping reflects the transition of the normal shock to the oblique shock. The results also show that the Euler equations can give the damping coefficients with a comparable accuracy.

중력이 작용하는 압축성유체를 고려함에 있어, 밀도성층 뿐만 아니라 엔트로피 성층의 고려도 중요하다. 본 연구에서 압축성격자볼츠만 유체모델을 이용한 차분계산법을 이용하여 2차원 채널에서 성층류의 전형적인 형상인 선택취수현상을 시뮬레이션 하였으며, 본 모델의 유효성을 확인하였다. 또한 비점성, 비압축성유체의 선택취수흐름과의 차이에 관해서 압축성의 관점에서 고찰하였다.

An adaptive Cartesian grid method having the best elements of structured, unstructured, and Cartesian grids is developed to solve the steady two-dimensional Euler equations. The solver is based on a cell-centered finite-volume method with Roe's flux-difference splitting and implicit point Gauss-seidel time integration method. Calculations of several compressible flows are carried out to show the efficiency of the developed computer code. The results were generally in good agreements with existing data in the literature and the developed code has the good ability to capture important feature of the flows.

This paper deals with the development of shock stable scheme that is free from shock instability. Roe's FDS is known to preserve good accuracy but to suffer from shock instability, i.e. the carbuncle phenomenon. As the first step toward the shock stable scheme, Roe's FDS is compared with HLLE scheme to identify the source of shock instability. Then control function f is introduced into the pressure term in Roe's FDS to cure shock instability. Various numerical tests concerned with shock instability are performed to demonstrate the shock stability of the proposed scheme.

An Eulerian-Lagrangian method, so called immersed boundary method, is used for analysing viscous flow around arbitrary bodies, where governing equations are discretized on a regular grid by using a finite volume method. To improve the accuracy of flow near body boundaries, a second-order accurate interpolation scheme is used and a level-set based grid deformation method is presented to construct the adaptive grids around body boundaries. The present scheme is used to simulate steady flow around a semicircular cylinder mounted on the bottom of flow domain and calculated results are validated by results of a body fitted grid method. Finally, present method is applied to a complex flow around multi body and the usefulness is checked by investigating calculated results.

The study of nonlinear gas transport in rarefied condition or associated with the microscale length of the geometry has emerged as an interesting topic in recent years. Along with the DSMC method, several fluid dynamic models that come under the general category of the moment method or the Chapman-Enskog method have been used for this type of problem. In the present study, on the basis of Eu's generalized hydrodynamics, a computational model for diatomic gases is proposed. The preliminary result indicates that the bulk viscosity plays a considerable role in fundamental flow problems such as the shock structure and shear flow. The general properties of the constitutive equations are obtained through a simple mathematical analysis. With an iterative computational algorithm of the constitutive equations, numerical solutions for the multi-dimensional problem can be obtained.

The impactive Pressure wave induced by pyro valve in the pressurization system of KSR-III may result in fatal failure or malfunction of valves, so it is important to know the pressure wave characteristic in the pressurization system. An unsteady flow analysis in the simplified pressurization system was peformed by Finite difference method.

An accelerating flow field through a compressor cascade is studied numerically by unsteady computational simulation. The two-dimensional Navier-Stokes equations for compressible flow is used for the study of unsteady high incidence angle flow, with preconditioning scheme to cover the wide range of Mach number and $\kappa-\omega$ model for the turbulent viscous flow analysis. A DCA(double circular arc) compressor blade is accelerated artificially in this study to understand the unsteady effect by comparing the present results with the existing steady-state experimental and computational results. Also, the accelerating flow field during the starting phase of gas turbine is studied with actual experimental data for the understanding of flow field and performance characteristics at off-design condition.

In this paper, the preconditioned multistage time stepping methods which are popular multigrid smoothers is studied for the compressible flow calculations. Fourier analysis on the local time stepping and block-Jacobi preconditioned residual operators is performed using the linearized 2-D Navier-Stokes equations. It fumed out that block-Jacobi preconditioner has better performance in eigenvalue clustering. They are implemented in the 2-D compressible Euler and Wavier-Stokes calculations with multigrid methods to verify that the block-Jacobi preconditioned multistage time stepping shows better performance in convergence acceleration.

Viscous solutions of supersonic jet impinging on a flat plate normal to the flow are simulated using three-dimensional Navier-Stokes solver. The jet impinging flow structure exhibits such complex nature as shock shell, plate shock and Mach disk depending on the flow parameters. Among others, the dominant parameters are the ratio of the nozzle exit pressure to the ambient pressure and the distance between the nozzle exit plane and the impinging plane. In the present study, the nozzle contour and the pressure ratio are held fixed, while the jet impinging distance is varied to illuminate the characteristics of the jet plume with the distance. As the plate is placed close to the nozzle at 3D high, the computed wall pressure at or near the jet center oscillates with large amplitude with respect to the mean value. Here D is the nozzle exit diameter. The amplitude of wall pressure fluctuations subsides as the distance increases, but the maximum pressure level at the plate is achieved when the distance is about 4D high. The frequency of the wall pressure is estimated at 6.0 kHz, 9.3 kHz, and 10.0 kHz as the impinging distance varies from 3D, 4D, to 6D, respectively.

Characteristics of high temperature rocket nozzle flow is discussed along with the aspects of computational analysis. Three methods of nozzle flow analysis, frozen-equilibrium, shifting-equilibrium and non-equilibrium approaches, were discussed those were coupled with the methods of computational fluid dynamics. A chemical equilibrium code developed for the analysis of general hydrocarbon fuel was coupled with three approaches of nozzle flow analysis, and a test was made for a bell nozzle at typical operation condition. As a results, the characteristics of the approaches were discussed in aspects of rocket performance, thermal analysis and computational efficiency.

Parallel performance of a Myrinet based PC-cluster was tested and compared with a conventional Fast-Ethernet system. A preconditioned Navier-Stokes code was parallelized with domain decomposition technique, and used for the parallel performance test. Speed-up ratio was examined as a major performance parameter depending on the number of processor and the network topology. As was expected, Myrinet system shows a superior parallel performance to the Fast-Ethernet system even with a single network adpater for a dual processor SMP machine. A test for the dependency on problem size also shows that network communication speed is a crucial factor for parallelized computational fluid dynamics analysis and the Myrinet system is a plausible candidate for high performance parallel computing system.

When caught in a fire inside a building or a tunnel the generated smoke is the main cause of the bad visibility, which makes difficult for a person to find escape route. Therefore it is required to visualize the simulated results of smoke realistically from a viewpoint of a person caught in a fire. In the present study, demonstrated is a CFD post-processor which can visualize the objects through smoke from the results of CFD fire simulation.

At the end of 1999, the TeraCluster Project in the KISTI Supercomputing Center was initiated to explore the possibility of PC clusters as a scientific computing platform to replace the Cray T3E system in KISTI by 2002. Since actual performance of a computing system varies significantly for different architectures, representative in-house codes from major application fields were executed to evaluate the actual performance of systems with different combination of CPU, network and network topology. As an example of practical CFD(Computational Fluid Dynamics) simulations, the flow past the Onera-M6 wing and the flow past a infinite wing were simulated on a clusters of Linux and several other hardware environments.

Fire inside light railway transit(LRT) was simulated for various fire scenarios using CFD analysis in order to assess the safety of passengers. Considered in the present paper are the effects of locations of fire, a fire size, an operation of ventilation system and an opening time of door on evolution of temperature and smoke concentration inside the LRT. For fire simulation, fire cells releasing heat and smoke corresponding to fire size were positioned at the location of fire. From the results, it is seen that the ventilation system and the opening time of door were the most important factors on temperature and smoke concentration inside the LRT

For windshield defrosting, flow analysis of inner room(vehicle) and heat conduction on the windshield surface are undertaken. Simulation for defrosting enthalpy method is usedand verification of heat and fluid flow analysis for room is done in cavity flow. The defrosting process is three dimensional phenomena and phase is changing. The result of defrosting analysis are well presenting the phase change and these results offer basic design data for defrosting phenomena.

The flow past a circular cylinder forced to vibrate transversely is numerically simulated by solving the two-dimensional Wavier-Stokes equations modified by the vibration velocity of a circular cylinder at a Reynolds number of 164. The higher-order finite difference scheme is employed for the spatial discretization along with the second order Adams-Bashforth and the first order backward-Euler time integration. The calculated cylinder vibration frequency is between 0.60 and 1.30 times of the natural vortex-shedding frequency. The calculated oscillation amplitude extends to $25\%$ of the cylinder diameter and in the case of the lock-in region it is $60\%$. It is made clear that the cylinder oscillation has influence on the wake pattern, the time histories of the drag and lift forces, power spectral density and phase diagrams, etc. It is found that these results include both the periodic (lock-in) and the quasi-periodic (non-lock-in) state. The vortex shedding frequency equals the driving frequency in the lock-in region but is independent in the non-lock-in region. The mean drag and the maximum lift coefficient increase with the increase of the forcing amplitude in the lock-in state. The lock-in boundaries are also established from the present direct numerical simulation.

In this paper, non-reacting and reacting flowfields were computed using a preconditioned Navier-Stokes solver. The preconditioning technique of Merkle et al. and TVD scheme or Chakravarthy and Osher was employed and the results obtained using developed code have a good agreement with the previous results and experimental data. The preconditioned Wavier-Stokes equation set with low Reynolds number $\kappa-\epsilon$ equation and species continuity equations, are discretized with strongly implicit manner and time integrated with LU-SSOR scheme. For the purpose of treating unsteady problem the duel-time stepping scheme was employed. For the validation of the code in incompressible flow regime, steady driven square cavity flow was considered and calculation result shows reasonably good agreement with the result of incompressible code. Shock wave/boundary layer interaction problem was considered to show the shock capturing performance of preconditioned-TVD scheme. To validate unsteady flow, acoustic oscillation problem was calculated, and supersonic premix flame of $H_2$-air reaction problem which is calculated with turbulence model, 9-species/18-reaction step reaction model, shows reasonable agreement with the previous results. As a result, the preconditioning method has an advantage to calculate incompressible and compressible flow through one code and preconditioned solver easily developed from standard compressible code with minor efforts. But additional computational time and computer memory is required due to preconditioning matrix.

In this paper, the convergence characteristics of preconditioned multigrid methods are investigated. The preconditioning method is introduced to reduce the condition number of discrete governing equations. 6 preconditioners including a point, line and diagonalized line solvers are implemented and applied to 2-dimensional inviscid flow problems. Theoretical fourier analyses and numerical results are presented for the preconditioners.

Any circumferential variations of the impeller exit flow conditions influences on the volute inlet flow conditions. All these interactions are strongly coupled phd affect consequently the performance of centrifugal pumps. In this paper, a commercial CFD code, which solves three-dimensional quasi-steady Wavier-Stokes equations with an impeller/volute interaction, is used for the prediction of a centrifugal pump performance. The simplified model of an impeller/volute interaction requires affordable computing time and provides relevant results. As a result, detailed flow structures such as pressure rise, recovery and loss mechanism on the centrifugal pumps are obtained. Especially, hydraulic performances are compared between the case of impeller only and the case of impeller with volute configuration. In addition, pump performance at off-design operation are observed and discussed.

RANS 방정식의 수치해법을 사용해서 tractor와 pusher 방식의 포드 프로펠러 주위 점성유동을 계산함으로써 포드형 전동 추진기의 추진 특성을 파악하고자 하였다. 본 연구에 사용한 수치해석 방법을 검증하기 위하여 최근 국내에서 모형 실험이 수행된 포드 프로펠러 형상에 대해 수치계산을 수행하고, 계산으로부터 얻어진 tractor와 pusher 방식의 포드 프로펠러에 대한 단독 성능을 실험 결과와 비교하였다. 또한, 포드에 작용하는 압력 및 프로펠러 날개 주위의 유동을 분석함으로써, 포드가 프로펠러의 추진 성능에 미치는 영향과 tractor와 pusher 추진 방식의 특성을 살펴보았다.