Toshiyuki Iwamiya

Flow simulation of NAL experimental supersonic airplane/booster separation using overset unstructured grids

Abstract The overset unstructured grid method developed for multiple-body problems is applied to a flow simulation about an experimental supersonic airplane separation from a rocket booster. An unstructured grid around the rocket booster is overset on the stationary grid around the airplane and moves with time to simulate the separation process. Detailed components of the rocket booster are faithfully reproduced by the unstructured grid. This capability of the unstructured grid reduces the number of required overset grids and significantly simplifies the overset procedure. The computed result of the airplane/booster separation clearly simulates the complex reflection patterns of shock waves between two bodies during the separation process. The computed lift and pitching moment coefficients are compared with the wind tunnel results. The computational accuracy of the aerodynamic coefficients is improved by including the detailed components of the airplane and rocket.

Numerical Optimization of Fuselage Geometry to Modify Sonic-Boom Signature

A low-sonic-boom design method is developed by combining a three-dimensional Euler computational e uid dynamics code with a least-squares optimization technique. In this design method, the fuselage geometry of an aircraft is modie ed to minimize the pressure discrepancies between a target low-boom pressure signature and a calculated signature. The aircraft cone gurations that generate three types of low-boom pressure signatures, i.e., e attop type, ramp type, and hybrid type, are successfully designed by this method. It is shown that the sonic-boom intensity of the aircraft designed by linear theory is reduced and the e attop-type ground pressure signature is obtained by this method. The results of the study suggest that this method is a useful tool for low-boom design.

Development and achievement of NAL Numerical Wind Tunnel (NWT) for CFD computations

NAL Numerical Wind Tunnel (NWT) is a distributed memory parallel computer developed through joint research and development of NAL and Fujitsu. It is based on the analysis of CFD codes developed in NAL. The target performance is more than 100 times faster than VP400. In this paper, the parallel computation model employed in the development of the NWT is described. The specification and feature of the NWT and the NWT Fortran are discussed. Finally, performance evaluations and some applications are presented. We find that the target performance is attained.<<ETX>>

Towards understanding of confinement of gluons

Abstract We report large scale numerical simulation data for gluon propagators together with an analysis based on the generalized Feynman rules of an extended perturbation scheme. Zwanzigers stochastic gauge fixing algorithm is employed to limit the path integration inside the Gribov region. Obtained propagators do not show a simple physical pole behavior, rather complex-conjugate pairs of singularities in the p2 plain.

Wing design for supersonic transports using integral equation method

Abstract The applicability of an inverse problem solver systemized with other computational tools was demonstrated. The demonstrations have been conducted on the aerodynamic design of wings for two Japanese scaled experimental Supersonic Transport (SST) models. The first model is of a clean configuration which has no propulsion system while the second one has two big engines. The design has been primarily performed by the system of Computational Fluid Dynamics (CFD) tools, which are a Navier–Stokes equation solver, geometry defining software, an inverse problem solver and interface programs between each solver and another. Because of the challenging concept of the wing design for the SSTs, such as a Natural Laminar Flow wing, a new design method was needed. Then, an integrated CFD design system of residual–correction loop was developed utilizing the existing inverse problem solver which solved integral equations. Though the inverse problem solver was based on low order approximation of flow equations, the design system worked successfully for the first model in Navier–Stokes flows. As for the second one that was a fairly complicated design problem affected by a big propulsion system, the preliminary stage of the wing design finished. Through both results, it was shown that we attained synergy effect of each CFD tool by systemizing and the design system using the simple inverse problem was promising for complex three-dimensional design.

Fuselage Shape Optimization of a Wing-Body Configuration with Nacelles

An aerodynamic design tool that combines a computational e uid dynamics code with an optimization technique for a drag minimization is developed and applied to a fuselage shape optimization of a Mach 1.7 scaled supersonic experimental airplane. An airframe/nacelle integration is taken into consideration. The optimized fuselage is compared with a conventional axisymmetrical area-ruled fuselage designed by a lineartheory. The results indicate that a nonaxisymmetrical fuselage design concept with this optimization design tool iseffectivefor the reduction of pressuredrag,especially inthedesign ofan airplanethatgeneratesa strong interferencedrag between itsairframe and nacelles.

Monotonicity, resolvents and Yosida approximations of operators on Hilbert manifolds

Abstract Monotonicity, resolvents and Yosida approximations of operators on manifolds of class C 3 are considered. Strong convergence of difference approximations to differential equations is shown by using resolvents. Our motivation is to give an elementary step toward constructing nonlinear semigroup theory on infinite dimensional manifolds.

Simulation of the 3 dimensional cascade flow with numerical wind tunnel (NWT)

The NWT was reinforced to get 280 GFLOPS of theoretical peak performance with the addition of 26 PEs to the original 140 PEs. On a CFD simulation of jet engine compressor, we attained active performance speed of 111 GFLOPS using 160 PEs.

Computational Fluid Dynamics Evaluation of National Aerospace Laboratory Experimental Supersonic Airplane in Ascent

Aerodynamic coefe cients of the National Aerospace Laboratory experimental airplane piggybacked on a solid rocket booster in ascent are numerically evaluated by solving the Euler equations on unstructured tetrahedral grids. For accurate representation of the cone guration, a computational method has been developed. The method effectively links a computer-aided-design-based modeling to the efe cient surface meshing. It also enables quick addition/removal of the small components attached on the original model to investigate their aerodynamic effects. The computed lift coefe cients show good agreement with wind-tunnel data in all regions of e ight speed from transonic to supersonic e ow. It is made clear that a small component affects physical phenomena the e owe eld, and as a consequence, the total aerodynamic performance of the airplane is largely changed. It is concluded that the detailed representation of the original geometry is very important for accurate evaluation of the transonic lift coefe cients.

Gluon propagators and confinement

We present SU(3) gluon propagators calculated on 48*48*48*N_t lattices at beta=6.8 where N_t=64 (corresponding the confinement phase) and N_t=16 (deconfinement) with the bare gauge parameter,alpha, set to be 0.1. In order to avoid Gribov copies, we employ the stochastic gauge fixing algorithm. Gluon propagators show quite different behavior from those of massless gauge fields: (1) In the confinement phase, G(t) shows massless behavior at small and large t, while around 5<t<15 it behaves as massive particle, and (2) effective mass observed in G(z) becomes larger as z increases. (3) In the deconfinement phase, G(z) shows also massive behavior but effective mass is less than in the confinement case. In all cases, slope masses are increasing functions of t or z, which can not be understood as addtional physical poles.We present SU(3) gluon propagators calculated on 48*48*48*N_t lattices at beta=6.8 where N_t=64 (corresponding the confinement phase) and N_t=16 (deconfinement) with the bare gauge parameter,alpha, set to be 0.1. In order to avoid Gribov copies, we employ the stochastic gauge fixing algorithm. Gluon propagators show quite different behavior from those of massless gauge fields: (1) In the confinement phase, G(t) shows massless behavior at small and large t, while around 5<t<15 it behaves as massive particle, and (2) effective mass observed in G(z) becomes larger as z increases. (3) In the deconfinement phase, G(z) shows also massive behavior but effective mass is less than in the confinement case. In all cases, slope masses are increasing functions of t or z, which can not be understood as addtional physical poles.