Search

A supersonic jet into complex geometry and its accompanying acoustic field have been investigated by using Euler Equation/Linearized Euler Equation (LEE) hybrid methodology here. As a first step, we have examined its performance by computing traveling sinusoidal waves. Then, we found that the hybrid methodology achieves higher accuracy than the Euler computation when combined with a fifth-order WENO (Weighted Essentially Non-Oscillatory) scheme. Based on this knowledge, we have conducted both of fully third-order Euler/LEE and third-order-Euler/fifth-order-LEE hybrid computations. According to the results, the computed acoustic field is mainly composed of three waves having distinct frequencies: a high frequency wave (0.5 Hz) which seems to be caused by interferences from the wall, a medium frequency wave (2.2 Hz) corresponding to first-mode eigenfrequency of the jet exit geometry, and a low frequency wave (10 Hz). The computed pressure values are rather different between the two results, although these frequencies are not so much. This fact implies that the more we extend the region of fifth-order computation, the more the computed pressure values approach to physically correct ones in quantitative view., JAXA Special Publication, 宇宙航空研究開発機構特別資料

Detached-Eddy Simulation is applied to three airfoils with different stall types. The method combines the strong points of Reynolds-averaged Navier-Stokes and Large Eddy Simulation approaches. Spalart-Allmaras Model is basically used, which reduces to a RANS formulation near a solid surface and to a subgrid model away from the wall. LU-SGS implicit scheme is employed to solve the model in time. Compared with experiment data, it is found that stall angle can be reasonably predicted by the present method., JAXA Special Publication, 宇宙航空研究開発機構特別資料

A numerical analysis of compressible and viscous flow around the Orbital Reentry Experiment (OREX) has been carried out with the Roe's approximate Riemann solver and Monotone Upstream-centered Scheme for Conservation Law (MUSCL) type extrapolation. The calculation was performed at Mach number 7.1 and Reynolds number 6.518 x 10(exp 6)., Roeの近似リーマン解法とMUSCL型の外挿法を用いて、OREXのまわりの圧縮性･粘性流れの数値解法を行った。計算は、マッハ数7.1,レイノルズ数6.518×10{6}で行った。

Reflection and refraction of linear disturbance waves by an oblique shock wave is studied by a linear analysis. Several different cases are considered, when the incident plane wave is a fast acoustic, slow acoustic, entropy, or vorticity wave. Results show that 1) a critical angle of the wave incidence exists, beyond which the regular wave solution of the linear problem can not be realized, 2) strongest sound waves are generated behind the shock wave, if the incident wave strikes the shock wave at the critical angle, and 3) at large Mach numbers of the upstream flow, the sound amplification factor asymptotically increases as the Mach number squared, while at the critical incidence, it is increased as the Mach number cubed., JAXA Special Publication, 宇宙航空研究開発機構特別資料

Cartesian grid does not follow the contour of body, as in commonly used, body-fitted schemes. This grid makes grid generation easy and reduces generation time cycle when handling complex body geometry such as complete aircraft configuration. These are the main advantages over body-fitted schemes. However, in a simple Cartesian grid generation, when the body dimension is less than cell size (thin body), the separate fluid regions in the cell are mistakenly considered as one fluid region (multiple-flow-region problem), which produces erroneous result. In this paper a new cell cutting procedure is proposed, in which the cell is automatically split into separate sub-cells. In sharp edge of body similar treatment is also performed to eliminate truncation of the edge in a simple cell cutting procedure. The results show that the cell splitting algorithm proposed here is effective in preventing the errors while keeping the number of cells reasonably low., 通常用いられる物体適合スキームのように、矩形格子は物体の輪郭には従わない。この格子は、完全航空機形態のような複雑機体形状の取扱い時に、格子生成を容易にし、生成時間サイクルを短縮する。これらは機体適合法に優る主な利点である。しかしながら、単純矩形格子生成では、機体寸法がセルサイズより小さい(薄い機体)場合、セルの中にある別の流体領域を誤って1つの流体領域(多流域問題)として考えるため、誤りの結果を生ずる。本論文では、セルを自動的に個別のサブセルに分割する新しいセル切断手順を提案した。機体の鋭く尖ったエッジにおいても、単純なセル切断手順ではエッジが切り落とされるという欠点を除くために、同様の取り扱いをした。ここで提案のセル分割アルゴリズムは、セル数を適度に低く保ちながら、誤りを防ぐ効果があることを示した。

In this computation, an upwind method with MUSCL (Monotonic Upstream Schemes for Conservation Laws) type extrapolation was applied to nonequilibrium chemical reacting hypersonic flow around an Orbital Reentry Experiment (OREX) configuration. The flow solver employed in this computation is based on Steger and Warming's flux-vector splitting (FVS), and classified as a variation of the Advection Upstream Splitting Method (AUSM). It has been confirmed that the present scheme has many good characteristics in the ideal gas flow such as monotonicity for shock wave, robustness for expansion wave, and high resolution for contact discontinuity and shear layer. In order to extend the present scheme to a chemical reacting flow, Park's two temperature model for five neutral species (N, O, NO, O2, N2) were used with 17 chemical reaction, and the finite volume formulation was employed to discretize the axisymmetric Navier-Stokes equations., MUSCL(保存則のための単調な上流スキーム)型補外法を持つ風上法を軌道再突入実験(OREX)コンフィギュレーション回りの非平衡化学反応極超音速流に適用した。この計算に使った流れ解法は、Steger-Warmingの流束ベクトル分割法(FVS)を基にし、移流上流分割法(AUSM)の変形として分類される。現在のスキームは、衝撃波に対する単調性、膨張波に対するロバスト性および接触不連続や剪断層に対する高分解能のような理想気体流において多くの良い特徴を持っている。現在のスキームを化学反応流れに拡張するために、17の化学反応をもつ5つの中性種(N,O,NO,O2,N2)に対しParkの2温度モデルを使い、軸対象ナビエ・ストークス方程式と区別するために有限体積定式化を使った。

Recently CFD is commonly used in design, where high efficiency is required. The cartesian grid has an advantage over other grids because it is easy to generate grid. However, many grid cells are necessary to correctly capture various flow phenomena, which spoils the merit of the cartesian grid. In the present study, an LU-SGS method is employed for time integration to overcome this problem., Special Publication of National Aerospace Laboratory, 航空宇宙技術研究所特別資料

A hybrid explicit/implicit numerical scheme referred to as the CCG(Collins,Collela and Glaz) scheme was recently proposed by Collins, Collela, and Glaz for one-dimensional hyperbolic conservation laws. By suitable blending of an explicit second-order time marching scheme with an implicit scheme, this approach was made to posses the Max Norm Diminishing (MND) property for all Courant numbers. Having been manifested for linear equations, the CCG scheme, however, fails to maintain this property unconditionally for non-linear equations, so that it requires an additional CFL (Courant-Friedrichs-Lewy)-like restriction on the time step. In this paper it is shown how to remedy the shortcoming of the CCG scheme, and also proposed a new general approach to design unconditionally MND hybrid schemes for non-linear advection equations. Numerical experiments are carried out for calculating the Burgers equation on a highly non-uniform grid. Results of these calculations exhibit a certain advantage in accuracy and efficiency of the proposed hybrid scheme compared with both the conventional implicit and the second-order explicit schemes., CCG(Collins-Collela-Glaz)スキームと呼ばれるハイブリッド陽的-陰的高解像度法が1次元双曲型保存則のために、Collins、CollelaおよびGlazにより提案された。陰的解法スキームを陽的2次精度時間進行スキームと適当に混合することにより、この方法は、すべてのクーラン数に対してMND(最大ノルム低減)特性を持つようにつくられている。このCCGスキームは線形方程式では証明されているが、非線形方程式に対してはこの性質を無条件に維持する様になっていない。それゆえ、その時間ステップに付加的なCFL(Courant-Friedrichs-Lewy)類似の条件が必要となる。本稿では、このCCGスキームの欠点を修復する方法を示し、非線形移流方程式にたいしてMNDハイブリッドスキームを無条件に設計するための新しい一般的なアプローチを提案する。数値実験をバーガース方程式の高度に不均一な格子上での計算にたいして実施した。これらの計算の結果により、提案したハイブリッドスキームは通常の陰的スキームや2次精度陽的スキームに比べて、精度と効率においてある一定の優位さをもつことを示した。

Genetic algorithms (GAs) have been successfully applied to numerical optimization problems, unfortunately GAs still remain computationally expensive, and the high computational cost make the use thereof impractical in most Aerodynamic optimization problems. Computational cost reduction is undoubtedly thought to be a common problem of most Aerodynamic shape optimization. In this approach, Artificial Neural Network (ANN) is used for function approximation; a number of mathematical computations are performed on the approximated function to obtain a reduced model. This resulting mathematical model is used to locate the variables that affect most the cost function. This approach is first tested on analytical functions, then numerical experimentations are conducted to solve shape optimization problem for the design of a wing profile. For each evaluation required by the optimizer, the Navier-Stokes equations with the Baldwin-Lomax turbulence model are solved., JAXA Special Publication, 宇宙航空研究開発機構特別資料

A grid adaptation method using anisotropic Cartesian grid in three dimensions has been developed to improve the efficiency of an existing Cartesian grid adaptation by reducing the total number of cells needed to resolve flow features. The method is capable of coarsening and refining a grid in such a way that the cell aspect ratio can take an arbitrary value. Flow computations with the present method for a supersonic flow around a cylinder and a transonic flow around an ONERA M6 wing are presented here, which show good agreement with experiment. For the case of cylinder, the present method produces about 67 times fewer cells than the isotropic approach. This dramatic saving is due to the ability of the three-dimensional code employed here to treat cells without any limitation on their aspect ratio. For the case of ONERA M6, the ratio of saving is about 2.5. Though the amount of memory per cell is higher than that of the isotropic approach, the overall memory requirement becomes lower due to reduction in the number of cells. This means that the present method can make flow computation more efficient., 流れ特性の解明に必要なセル全体の数を減らし、現行のデカルト格子の適用効率を改善するため、3次元におけるデカルト格子を使い格子適用法を改善してきた。この方法ではセルのアスペクト比が任意の値を取る事ができるように格子を粗くしたり、細かくする事ができる。この方法を用いて円筒の周りの超音速流とONERA M6翼の周りの遷音速流に対する流れの計算を示した。これらは実験と良く一致した。この方法ではセルの数は等方性の方法より67分の1と少ないセルの数であった。この劇的な節約はここでスペクト比に制限を加える事なくセルを取り扱うために採用した3次元コードの能力に依る。ONERA M6の場合は節約の比率は約2.6であった。セル当たりの記憶量は等方性の方法に比べ大きいが、全体として必要な記憶量はセルの数が減少しているので少ない。この事はこの方法により流れの計算をより効率良くできる事を意味している。

A numerical scheme to treat the problem of fluid-structure coupling has been developed in this study. A high order scheme based on the finite volume method (FVM) is used for fluid calculation, as well as the finite element method (FEM) for structure calculation in order to make more detailed analysis of flutter phenomena. The present method has been applied to the problem of panel flutter. For the supersonic case of M = 1.2, the flutter shows a rapid increase in amplitude into a limit cycle oscillation with a first mode. From the comparison between Euler and Navier-Stokes computation, it is clear that a boundary layer delays the growth of the flutter oscillation. For the subsonic case of M = 0.95, a static, upwardly divergent solution has been obtained, where vortices are produced due to the fluid-structure interaction in the Navier-Stokes computation., Special Publication of National Aerospace Laboratory, 航空宇宙技術研究所特別資料

A Computational Fluid Dynamics (CFD) technique is employed to study hypersonic high enthalpy air flow around blunt bodies at a range of enthalpies relevant to suborbital flight speeds of Aeroassisted Orbital Transfer Vehicles (ASTVs). The method uses the two-temperature model of Park for the description of thermo-chemical nonequilibrium processes in high temperature air and solves the full Navier-Stokes equations for multicomponent reacting gas mixture in finite volume formulation on a grid of arbitrary structure. The calculations performed in this work simulate in detail the experiments carried out recently at the KHI (Kawasaki Heavy Industries) (Japan), and at the DLR (Germany), where heat flux distribution over a sphere and a capsule-like blunt cone model was measured at several freestream conditions related to the range of enthalpies mentioned above. The main emphasis in this paper is given to comparison between numerical and experimental results in order to verify adequacy of data in terms of heat flux distribution predicted by the CFD technique for suborbital ASTV flight range., 空力利用軌道間輸送機(ASTVs)の軌道に乗らない飛行速度に相当するエンタルピーの範囲で、鈍頭物体回りの極超音速高エンタルピー空気流れを研究するために計算流体力学(CFD)技術を採用した。この方法は、高温空気中で熱化学非平衡過程を記述するためにParkの2温度モデルを使用し、任意構造のグリッドの有限体積定式化において多成分反応ガス混合物に対し完全ナビエ・ストークス方程式を解いた。本研究で実施された計算は、日本の川崎重工業(KHI)およびドイツのDLR(ドイツ航空宇宙研究所)で最近実施された実験を詳細にシミュレートした。ここで球体およびカプセル状鈍頭モデル上の熱流束分布が上述したエンタルピーの範囲に関連した幾つかの自由流れ状態で計測された。この論文の主な強調点は、軌道に乗らないASTV飛行範囲をCFD技術により予想した熱流束分布によってデータの精度を実証するために、数値および実験データの比較を行ったことである。

Although the flow rate through many holes on a transonic wind tunnel wall is very small， this flow makes a significant influence on the wind tunnel test results． However there are less data which show the relationship between the flow rate and the differential pressure through holes at this flow conditions． In this study， the relationship between the flow rate and the differential pressure through a hole on a transonic wind tunnel wall is experimentally investigated． In the case that the flow rate and the differential pressure are very small， the results show that their relationship is linear． Furthermore the slope of the line measures up to the results of CFD．, Original contains color illustrations, Meeting Information: 42nd Fluid Dynamics Conference / Aerospace Numerical Simulation Symposium 2010 (June 24-25, 2010. Yonago Convention Center BiG SHiP), Yonago, Tottori Japan, 形態: カラー図版あり, 会議情報: 第42回流体力学講演会/航空宇宙数値シミュレーション技術シンポジウム2010 (2010年6月24日-25日. 米子コンベンションセンター BiG SHiP)

Acoustic loads are the principal source of structural vibration and internal noise during launch. Therefore, it is important to predict the acoustic loads on spacecraft such as a rocket. Conventionally, the prediction has been made by empirical methods. In these methods shielding and reflection are not considered. To solve these problems, we have been developing an Euler/LEE(Linearized Euler Equation) hybrid code and applying it to the acoustic evaluation of H-IIA's launch pad. In this report, we explain how to use the LEE option of the hybrid code., ロケット打ち上げ時に生じる構造振動や内部騒音の主因となる音響荷重を予測することは重要である。この予測は、従来、経験的手法によって行われてきたが、そうした手法では音響伝播過程における遮蔽や反射が考慮されていない。こうした問題を解決するために、我々はEuler/LEEコードを開発し、H-IIAの打ち上げ射場の音響評価に応用してきた。本報告では、第1報に引き続き当コードのLEEオプションの使用方法を解説する。, 形態: カラー図版あり

In the present study, motions of two small solid particles, which are lifted by the shock-induced flow behind a propagating shock wave, are numerically simulated to investigate the dynamic characteristics from the viewpoint of direct and indirect interactions between the two particles. The particle Reynolds number is 6 000, and the time period considered ranges from the time when the shock passes the particles to the time when the boundary layer thickness grows to the same order as the particle size. Three cases are considered for their initial arrangement on the wall. In the first case, they are placed tandem, aligned with the direction of shock propagation. Results show that they are lifted up, keeping contact each other. In the second case, they are seperated by three times as large as the particle size. As a result, the rear particle is lifted higher because the flow before the rear particle has been changed by the front particle. In the third case, they are placed in the spanwise direction. They are also lifted by the shock-induced flow, increasing the separation distance between them due to repulsive fluid force. In these three cases, the average height of two particles becomes larger than that in the case of one particle alone, which was studied in the previous work.

A method for performing aeroacoustic simulation, where the acoustic field is split from the flow field, is applied to the flow around an automobile rear-view mirror placed on a large solid plane. The flow can be obtained by performing large eddy simulation for the incompressible flow equations on colocated grids, while the acoustic equations are defined by the difference between the compressible and the incompressible flow equations, and solved by the finite volume method with the fourth order WENO scheme. The non-reflecting boundary conditions with a perfectly matched layer are applied to the outer boundary. Calculated results show clear sound emissions, where acoustic streamlines as propagation path are depicted based on acoustic intensity vectors. In this paper, two kinds of mirror shape, i.e., the eases with and without a strong vortex at the tip of a rear-view mirror are compared. Calculated results show some differences between them in the acoustic field. The sound pressure related with the pressure frequency of the generated vortex is increased in all directions, though the propagation path dose not show a large difference except near part of the body surface.

A method for performing aeroacoustic simulation, where the acoustic field is computed by the equations split from the flow field, is applied to the acoustic field of an object on the large flat surface. The acoustic field is obtained by the equations defined by the difference between the compressible and the incompressible flow equations. The acoustic equations are solved by the finite volume method with the fourth order WENO scheme. A rectangular cylinder with the limited width and an automobile rear-view mirror are calculated and the acoustic fields are investigated. The aspects in the near fields are different, while the emission patterns in the far fields are similar. The result of the rear-view mirror shows that the fluctuations of the regions with low acoustic density located downstream both outside and inside the mirror cause the acoustic emission.