研究者データベース

研究者情報

マスター

アカウント(マスター)

  • 氏名

    寺島 洋史(テラシマ ヒロシ), テラシマ ヒロシ

所属(マスター)

  • 工学研究院 機械・宇宙航空工学部門 宇宙航空システム

所属(マスター)

  • 工学研究院 機械・宇宙航空工学部門 宇宙航空システム

独自項目

syllabus

  • 2021, 流体工学特論, Advanced Fluid Engineering, 修士課程, 工学院
  • 2021, 大学院共通授業科目(一般科目):自然科学・応用科学, Inter-Graduate School Classes(General Subject):Natural and Applied Sciences, 修士課程, 大学院共通科目, Fluid engineering, Theory of thermal-fluid mechanics, Computational fluid mechanics, Flow measurement & visualization, Food Science
  • 2021, 大学院共通授業科目(一般科目):自然科学・応用科学, Inter-Graduate School Classes(General Subject):Natural and Applied Sciences, 修士課程, 大学院共通科目, Aerodynamics, airfoil, wing, transonic flows, supersonic flows, high enthalpy flows, high-temperature gas dynamics
  • 2021, 飛翔体工学特論, Introduction to Flight Dynamics, 修士課程, 工学院, 空気力学,翼型,翼,遷音速流れ,超音速流れ,高エンタルピー流れ,高温気体力学
  • 2021, 流体工学特論, Advanced Fluid Engineering, 博士後期課程, 工学院
  • 2021, 飛翔体工学特論, Introduction to Flight Dynamics, 博士後期課程, 工学院, 空気力学,翼型,翼,遷音速流れ,超音速流れ,高エンタルピー流れ,高温気体力学
  • 2021, メカトロニクス実習, Experiment on Mechatronics, 学士課程, 工学部, メカニクス、電子回路、制御、マイクロコンピュータ
  • 2021, 流体工学, Fluid Engineering, 学士課程, 工学部, 圧縮性流れ,音波,衝撃波,完全流体の力学,物体に働く力,翼理論,層流,乱流,境界層,航空機,空力,遷音速流

researchmap

プロフィール情報

学位

  • 博士(工学)

プロフィール情報

  • 寺島, テラシマ
  • 洋史, ヒロシ
  • ID各種

    201501016995926911

対象リソース

業績リスト

研究キーワード

  • 数値流体力学   燃焼反応流体   超臨界流体   気液界面   空力弾性   

研究分野

  • 情報通信 / 計算科学
  • フロンティア(航空・船舶) / 航空宇宙工学
  • ものづくり技術(機械・電気電子・化学工学) / 熱工学
  • 社会基盤(土木・建築・防災) / 安全工学
  • 社会基盤(土木・建築・防災) / 社会システム工学

経歴

  • 2023年04月 - 現在 Stanford University Department of Mechanical Engineering Visiting associate professor
  • 2015年12月 - 現在 北海道大学 工学研究院 准教授
  • 2013年04月 - 2015年11月 東京大学 工学系研究科 特任准教授
  • 2011年04月 - 2013年03月 東京大学 工学系研究科 特任助教
  • 2009年04月 - 2011年03月 宇宙航空研究開発機構 宇宙航空プロジェクト研究員
  • 2007年04月 - 2009年03月 Worcester Polytechnic Institute Mechanical Engineering Department Research Scientist
  • 2005年04月 - 2007年03月 理化学研究所 研究員(2006年11月より客員研究員)
  • 2004年04月 - 2005年03月 東京大学 IML 機関研究員

学歴

  • 2000年04月 - 2004年03月   東京大学   航空宇宙工学専攻 博士課程
  • 1998年04月 - 2000年03月   東北大学   航空宇宙工学専攻 博士課程前期
  • 1994年04月 - 1998年03月   東北大学   工学部   機械航空工学科

受賞

  • 2016年 日本燃焼学会 平成28年度日本燃焼学会 奨励賞
     
    受賞者: 寺島洋史

論文

  • NAKATSUKASA Tomohito, AMANO Taishi, ARAKI Takahide, TERASHIMA Hiroshi, TSUBOI Nobuyuki, OZAWA Kohei
    Journal of Evolving Space Activities 1 n/a  2023年 [査読有り]
     
    The present study numerically investigates the flow fields of the twin- and tandem-jet in crossflow (JIC) under supercritical pressure. An ILES/RANS method is applied. In the present JIC, cryogenic hydrogen jets are vertically injected to warm temperature hydrogen crossflow through circular injector holes. The results show that the penetration of cryogenic jets into the crossflow is more enhanced with the tandem-jet than with the twin-jet. The higher jet penetration with the tandem jets is caused by a downstream jet behavior less influenced by the crossflow. Nevertheless, a mixedness analysis indicates that the mixing state at a downstream position is almost similar between the tandem and twin jets.
  • Toma Miyake, Hiroshi Terashima
    Lecture Notes in Electrical Engineering 912 607 - 621 2023年 
    This study numerically investigates the transonic flutter characteristics of a supercritical airfoil and discusses the similarities and differences compared to a conventional symmetric airfoil. A wing-section model with two-degree-of-freedom was adapted. An SC2-0610 supercritical airfoil was used for the investigation, and the results were compared with those of the NACA64A010 airfoil. Overall, similar flutter characteristics, such as transonic dip appearance, were observed between the two airfoils. The effect of the shape of the supercritical airfoil appears as the angle of attack effect; for example, the supercritical airfoil with an angle of attack of −2°(0°) shows qualitatively similar flutter characteristics to the NACA64A010 airfoil with an angle of attack of 0°(2°). Furthermore, in the present study, the supercritical airfoil exhibits unique flutter characteristics under high Mach number conditions because of the development of flow separation near the trailing edge on the lower surface. In particular, when the Mach number was 0.875, the flow reattachment near the trailing edge on the lower surface generated a high-pressure region, which induced a unique flutter oscillation under the second torsion mode.
  • Noah D. Oyeniran, Hiroshi Terashima
    Lecture Notes in Electrical Engineering 912 353 - 360 2023年 
    The present study computationally investigates the characteristics of unsteady aerodynamic forces around an oscillating airfoil in the transonic flow regime. Particular attention is paid to the role of shock wave and shock-induced boundary layer separation in unsteady aerodynamics. The Reynolds-averaged compressible Navier–Stokes equations are solved with SA and SST turbulence models. A well-known forced-pitching NACA64A010 airfoil experiment (Sanford and Gerald in AIAA J 11:1306–1312, 1980, [6]) is simulated, and the freestream Mach number, Reynolds number, and reduced frequency are set to 0.8, 1.2 × 107, and 0.2, respectively. The pitching airfoil with mean angles of attack of 0°, 2°, 4°, and 6° having the amplitude of 1° is parametrically simulated. It is observed that at the mean angle of attack of 0°, there is a phase delay of the lift coefficient against the angle of attack due to the delay of a shock wave movement over the airfoil surface. In contrast, a phase-advanced feature of unsteady aerodynamics appears in increasing the mean angle of attack (e.g., 4° and 6°). There is a phase transition of unsteady aerodynamics from the delay to advance, significantly caused by the appearance of shock-induced boundary-layer separation. The mean angle of attack around 3° may correspond to a transition condition between the phase-delayed and phase-advanced features. The present study demonstrated that the trend of the unsteady aerodynamic characteristics around the transonic oscillating airfoils largely changes with the mean angle of attack. The shock wave, the shock-induced separation, and their interaction play a crucial role in determining the unsteady aerodynamics such as the phase-delayed or the phase-advanced features.
  • Noah D. Oyeniran, Toma Miyake, Hiroshi Terashima, Ryoto Seki, Keiichi Ishiko, Taku Nonomura
    AIAA Journal 2022年10月07日 [査読有り]
  • Toru Ota, Hiroshi Terashima, Nobuyuki Oshima
    AIAA Science and Technology Forum and Exposition, AIAA SciTech Forum 2022 2022年 
    Combustion instabilities are a significant issue in combustion engine developments and operations. The present study numerically investigates combustion instabilities that happen in a high-pressure H2/O2 rocket-type combustor. The effects of fuel temperature on combustion instability are discussed. The compressible Navier-Stokes equations are solved with a detailed chemical mechanism, and a two-dimensional planar combustor model with a single injector is used for investigation. The simulation result demonstrates that the fuel injection temperature significantly affects the combustion instability feature, such as acoustic wave development. While the fuel injection with a room temperature results in stable combustion, the low-temperature fuel injection causes severe pressure oscillations with large amplitudes. The frequency analysis confirms that the pressure oscillation occurs in the first longitudinal mode of the combustor, i.e., the thermoacoustic combustion instability happens. The pressure (acoustic) wave development is caused by the heat release of unburned gas that remained in the near-field of the injector. For lower-temperature injection cases, a relatively large amount of unburned gas exists due to the delay of combustion, and thus the potential of pressure wave development in coupling with the heat release becomes higher. The development process may correspond to the deflagration-to-detonation process.
  • Hiroshi Terashima, Mitsuo Koshi
    Proceedings of the Combustion Institute 2022年 [査読有り]
     
    The present study discusses computational fluid dynamics (CFD) modeling for combustion flows under high-pressure conditions, including supercritical pressure conditions. The real-fluid effects are considered in terms of thermodynamic properties, transport properties, and chemical kinetics. In the present model, the real-fluid effect on chemical kinetics is introduced via a modified equilibrium constant derived using the Gibbs free energy variation of chemical reactions and the fugacity. The results obtained with the present model are compared with available experimental data of high-pressure premixed H2/O2 propagating flames diluted by Ar or He. We demonstrate in the H2/O2/Ar flame case that the real-fluid model provides a better prediction accuracy for the negative pressure dependence of the mass burning rate compared to the ideal-gas model. The improved prediction accuracy is primarily attributed to the proper estimation of thermodynamic properties such as unburnt-gas enthalpy via an appropriate equation of state and a departure function. The negative pressure dependence of unburnt-gas enthalpy of the H2/O2/Ar mixture with the real-fluid model significantly affects the flame speed prediction under high-pressure conditions. On the other hand, although the H2/O2/He mixture shows a positive pressure dependence of enthalpy, differences in the mass burning rate between the real-fluid and ideal-gas models are not significant for the H2/O2/He flame case. In the He-diluted case, the real- fluid effect is undermined owing to the low density of the H2/O2/He mixture. Thus, the real-fluid effect appears differently in the prediction of propagating flames depending on the species composition and thermodynamic conditions. The present study suggests that the positive or negative pressure dependence of enthalpy (i.e., the isothermal Joule-Thomson coefficient) is a metric to identify the real-fluid effects that appear.
  • Daiki Muto, Hiroshi Terashima, Takahide Araki, Nobuyuki Tsuboi
    Journal of Propulsion and Power 38 4 581 - 591 2022年 [査読有り]
     
    Effects of recess lengths on coaxial cryogenic jets under a supercritical pressure are numerically studied. A hybrid large-eddy simulation/Reynolds-averaged Navier–Stokes methodology is applied to capture unsteady jet behaviors of recessed injectors. For the injector with a moderate recess length, the dense-core length of the inner jet is shorter than that with a nonrecessed injector, demonstrating that recessing improves the mixing. The flowfields show that the vortex structures in an inner mixing layer are strongly developed within the recessed region, and the entrainment of the outer jet into the inner jet is strengthened. However, a further increase in the recess length turns to the increase of the dense-core length and deteriorates the inner jet decay along the centerline. This deterioration is caused by the separation of the outer jet flow on the wall in the recessed region. The turbulence generation in the outer mixing layer is suppressed because of the separation, resulting in less enhancement of the mixing in the downstream region. Therefore, the study demonstrates that there is an approximate recess length for optimum jet mixing. Finally, a correlation relationship is provided between the dense-core length and the momentum flux ratio from a comprehensive comparison of existing experimental and numerical data.
  • Takahisa Nogawa, Hiroshi Terashima
    Combustion Science and Technology 194 7 1433 - 1457 2022年 [査読有り]
     
    This study numerically investigates the effects of stratified temperature distributions on the end-gas combustion mode in a constant volume reactor. The initial temperature in the reactor was globally stratified with linear gradients. The effects of negative temperature coefficient (NTC) characteristics are addressed through a comparison between the results of a non-NTC fuel and an NTC fuel. The compressible Navier–Stokes equations are solved with detailed chemistry. The results show that the addition of large temperature gradients such as −10 or −5 K/cm can prevent the pressure wave development associated with end-gas autoignition, and the knocking intensity is significantly reduced. In contrast, small temperature gradients such as −1 K/cm lead to the generation of the developing detonation mode and thus to large knocking intensities. In the NTC fuel case, the knocking intensity becomes relatively large because the spatial gradient of the ignition delay time decreases even with the addition of large temperature gradients, especially in the NTC regime. Based on all the results, a correlation is presented between the initial temperature and the reaction front speed (the spatial gradient of the ignition delay time) for the prediction of end-gas combustion mode. The correlation suggests that the possibility of preventing large knocking intensities by temperature gradients is reduced under high-pressure conditions because of the increase in the reaction front speed. Consequently, this study suggests that very large temperature gradients would be required under high-pressure conditions to prevent large knocking intensities under the concept of thermal stratification.
  • Shun MURAKAMI, Hiroshi TERASHIMA, Nobuyuki OSHIMA
    TRANSACTIONS OF THE JAPAN SOCIETY FOR AERONAUTICAL AND SPACE SCIENCES, AEROSPACE TECHNOLOGY JAPAN 19 2 176 - 185 2021年 [査読有り]
  • Hiroshi Terashima, Yutaka Hanada, Soshi Kawai
    Proceedings of the Combustion Institute 38 2 2119 - 2126 2021年01月 [査読有り]
     
    The present study proposes a localized thickened flame (LTF) model for the accurate prediction of flame propagation and autoignition timing. The unresolved-scale terms appeared in spatially-filtered governing equations due to thin flame structures are constructed under a physical constraint in which laminar flame speed is maintained. A high-order derivative is introduced to dynamically localize the effects of the LTF in the regions of unresolved propagating flame. The model is also designed such that the thickened flame is resolved by the same number of grid points for any grid size used. Therefore, a user-specified constant in the model does not need to be adjusted depending on the employed grid size. Laminar flame propagation problems are used to validate the performance of the proposed LTF model and determine the appropriate value of the user-specified constant. The results using a one-dimensional constant-volume reactor demonstrate that the LTF successfully captures the accurate flame propagation behaviors under elevated pressure conditions, while not affecting the end-gas autoignition timing, even on relatively coarse grid resolutions. The high-order derivative in the LTF serves as a dynamic parameter for detecting the thinning flame under elevated pressure conditions.
  • Hiroshi Terashima, Hisashi Nakamura, Akira Matsugi, Mitsuo Koshi
    Combustion and Flame 223 181 - 191 2021年01月 [査読有り]
     
    This study highlights the importance of heat release rate in low-temperature oxidation (LTO) on non-uniform end-gas autoignition and strong pressure wave generation, which are substantially relevant to knocking combustion. The simulations are conducted using the compressible Navier–Stokes equations with detailed transport and chemical kinetics models in a one-dimensional constant-volume reactor. Four fuel/air stoichiometric mixtures, n-butane, i-octane, n-heptane, and dimethyl ether (DME)/air mixtures, are simulated. The results show that larger knocking intensities are produced with n-heptane and DME in their negative temperature coefficient (NTC) regimes because of the stronger non-uniformity of end-gas autoignition. The non-uniformity of end-gas autoignition is enhanced by a pressure wave disturbance that is caused by the rapid temperature rise of the end-gas region in LTO. In particular, the high heat release rate with the DME/air mixture generates a distinct pressure wave disturbance in the reactor, which considerably enhances the non-uniformity of end-gas autoignition through the reflection of the wave at the wall. In contrast, the heat release rate in the n-heptane case is milder than that in the DME case, and therefore, the knocking intensity in the n-heptane case is smaller compared to that of DME due to less enhancement of the non-uniform end-gas autoignition. No large knocking intensities are produced with n-butane and i-octane, which have weak NTC, because of the absence of a temperature rise in LTO. Thus, this study concludes that the high heat release rate in LTO and the generated pressure wave disturbance play a significant role in the generation of large knocking intensities through the enhancement of non-uniform end-gas autoignition.
  • Terashima, Hiroshi, Matsugi, Akira, Koshi, Mitsuo
    COMBUSTION AND FLAME 203 204 - 216 2019年05月 [査読有り][通常論文]
     
    The present study addresses effects of pressure wave disturbance on end-gas autoignition behavior and subsequent pressure wave development, which are of particular interest for understanding a fundamental process of knocking combustion in spark-assisted ignition engines. Such pressure wave disturbance is initiated by a compression wave generated from spark-like ignition and continuously exists in a combustion chamber. The investigation is conducted through a direct numerical simulation with a 1-D constant volume reactor, where the compressible Navier-Stokes equations are solved with detailed reaction mechanisms of n-C7H16 and n-C4H10/air mixtures. The width of an ignition kernel parametrically changes in order to control the strength of an initial compression wave and thus pressure wave disturbance. The result showed that stronger compression waves with larger kernel widths enhance the progress of the autoignition process at the wall through the transient increase in pressure and temperature due to the wave reflection. Consequently, stronger pressure waves such as a developing detonation are generated with the enhanced inhomogeneous end-gas autoignition starting from the wall. In contrast, weak pressure wave disturbance produces no strong pressure waves due to the lower inhomogeneity of autoignition timing in an end-gas region. An analysis with temperature variation in an end-gas region indicates that cool flame generation helps maintain the superiority of the wall for earlier autoignition through the heat release at an earlier stage. Therefore, a stronger pressure wave associated with inhomogeneous end-gas autoignition is likely generated around 650 K in the case of an n-C7H16/air mixture, which corresponds to a distinct peak generation of knocking intensity in the negative temperature coefficient regime. For lower temperature conditions such as 500 K, the end-gas autoignition first takes place in a region away from the wall. This is because the initially induced temperature inhomogeneity in the end-gas region is considerably weakened due to wave interactions and dissipation effects during longer ignition delay times. Thus, multiple end-gas points including the wall can become a preferred point for earlier end-gas autoignition, resulting in a variation of end-gas autoignition locations. The result for an n-C4H10 air mixture is finally shown and the difference from that for the n-C7H16/air mixture is highlighted. (C) 2019 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
  • Shun Murakami, Hiroshi Terashima, Nobuyuki Oshima
    AIAA Scitech 2019 Forum 2019年 [査読無し]
     
    A computational study is performed for exploring flow and flame dynamics of a high-pressure hydrogen/oxygen coflow jet with the effects of post thickness and moment flux ratio. A two-dimensional model with a splitter plate, which represents a post configuration of an injector of rocket engines, is adapted to fully resolve the combustion flow field. The compressible Navier-Stokes equations are solved with a detailed chemical kinetic mechanism in a manner of direct numerical simulation. The result shows that the post thickness largely affects the temperature distribution in a recirculation region established behind the post. The temperature distribution is determined with the amount of incoming high-temperature combustion gas and unburned hydrogen gas, which significantly changes with the post thickness. The effect of the momentum flux ratio clearly appears in the case of thicker post configuration, while in the case of thinner post configuration no major differences are identified for all the momentum flux ratio. The study shows a tendency that thicker post geometries with smaller J numbers provide lower temperature fields in the recirculation region behind the post, thus preliminarily indicating some difficulty of maintaining a flame anchoring in the recirculation region.
  • Mahiro Ooe, Hiroshi Terashima, Jun Hayashi, Fumiteru Akamatsu, Nobuyuki Oshima
    12th Asia-Pacific Conference on Combustion, ASPACC 2019 2019年 [査読無し]
     
    Ammonia is regarded as one of alternative fuels because of no CO2 emission during the combustion process. In this study, we numerically investigate details of flame holding/extinguishing mechanisms for a laminar non-premixed burner flame. The compressible Navier-Stokes equations with a detailed chemical kinetics of ammonia are applied in a manner of direct numerical simulation. The present result shows that increasing the rim thickness of a burner helps extend flame-holding conditions through the formation of larger recirculation region established behind the burner rim. However, further increase of the rim thickness eventually provides an unfavorable effect because of the decrease of heat release rate and the increase of heat loss to the rim wall. Thus, there exists an optimum rim thickness for increasing flame-holding capability.
  • Daiki Muto, Hiroshi Terashima, Nobuyuki Tsuboi
    Transactions of the Japan Society for Aeronautical and Space Sciences 62 4 203 - 212 2019年01月 [査読有り][通常論文]
     
    © 2019 The Japan Society for Aeronautical and Space Sciences. The effects of a recess on co-flowing planar jets under supercritical pressure are numerically studied. Two-dimensional hybrid LES/RANS simulations are performed in a wide range of recess lengths and injection momentum flux ratios, which are important design parameters for liquid rocket engine injectors. The present study showed that confinement effects of the near-injector flowfield by applying a recess, suppress outer jet spreading and thus enhances the penetration of the outer jet flow into the inner jet. The enhanced penetration of the outer jet flow results in the appearance of flapping motions in the inner jet. Low-frequency oscillations corresponding to the flapping motions clearly appear in the case of recessed injectors. Moreover, the confinement effects promote interactions between vortex structures resulting in vortex breakdown. Consequently, the inner-jet length is shortened, indicating an improvement in mixing when a recess is applied. The present results also show that the inner-jet length deceases as the recess length increases, and the effects of a recess remarkably appear at higher momentum flux ratios. This is explained by the vortices generated behind the post lip that is strengthened as the result of increased velocity ratio.
  • A. Razmi, M. Taeibi-Rahni, H. R. Massah, H. Terashima, H. Moezzi
    Journal of Applied Fluid Mechanics 12 2 631 - 645 2019年01月 [査読有り][通常論文]
     
    © 2019 Isfahan University of Technology. A front tracking/ghost fluid method was used to simulate fluid interfaces in a shock-bubble interaction problem. The method captures fluid interfaces, using explicit front-tracking and defines interface conditions, using the ghost-fluid method. In order to demonstrate the accuracy and the capability tracking of the approach used, an air-helium and anair-R22 shock-bubble interaction cases were simulated. The computational results were compared with reliable experimental and computational studies, showing close agreements.
  • D Deb, MA Uddin, H Terashima, N Oshima
    Journal of Scientific Research 10 2 117 - 131 2018年 [査読有り][通常論文]
  • Hiroumi Tani, Hiroshi Terashima, Yu Daimon, Mitsuo Koshi, Ryoichi Kurose
    COMBUSTION SCIENCE AND TECHNOLOGY 190 3 515 - 533 2018年 [査読有り][通常論文]
     
    Unsteady simulations of hydrazine (N2H4) sprays in nitrogen tetroxide (NTO, NO2-N2O4) streams were conducted to explore the hypergolic combustion in bipropellant thrusters. The Navier-Stokes equations were solved using a detailed chemical kinetics mechanism and dispersed droplets were modeled through direct numerical simulations. Auto-ignition occurred when the sum of the heat transfer from the ambient gas and the heat release from hydrogen abstraction reactions exceeded the latent heat of the droplets. Although the evaporation of the droplets was enhanced as the droplet size decreased, the ignition delay time increased due to the lower temperatures of the mixtures of the N2H4 vapor and nitrogen tetroxide. After the flames reached a steady state, a double flame structure appeared, comprised of outer diffusion and inner decomposition flames. The inner decomposition flame and N2H4 vapor flow exhibited a sinusoidal behavior at a certain droplet size. This behavior was initiated by the locally expanded decomposition gases and developed by the supply of N2H4 droplets to the decomposition gases at relatively high temperatures. In cases of larger and smaller droplet sizes, the sinusoidal behavior was not significant due to less evaporation of the N2H4 droplets and a lower temperature of the N2H4 vapor, respectively. The sinusoidal behavior of the decomposition flames enhanced the mixing and reactions of the fuel components (i.e., N2H4, NH3, and H-2). The present study demonstrated a large impact of droplet size on flame dynamics, suggesting that a fine spray is not always better for hypergolic propellant combustion to consume the fuel components quickly.
  • Ohashi Tatsushi, Takahashi Yusuke, Terashima Hiroshi, Oshima Nobuyuki
    JOURNAL OF FLUID SCIENCE AND TECHNOLOGY 13 3 2018年 [査読有り][通常論文]
     
    An inflatable membrane reentry vehicle has been developed as one of the innovative reentry technologies. A suborbital reentry demonstration using a sounding rocket was carried out in 2012. Contrary to the result of a preliminary study, the vehicle always had an angle of attack (AoA) during its reentry. In addition, the amplitude of AoA gradually increased as altitude decreased, and the vehicle rotated vertically under Mach number of 0.1 (M0.1). As a first step to clarify the cause of attitude instability and vertical rotation, the aerodynamic characteristics, that concern static stability, are numerically investigated. Numerical simulations were carried out for the cases of Mach 0.9 (M0.9), 0.6 (M0.6), 0.3 (M0.3), and 0.1 (M0.1) and pitching moment coefficients (CM) were obtained. Analysis software "RG-FaSTAR" for M0.9, and "FrontFlow/red" for M0.6, M0.3 and M0.1, are used, respectively. Large eddy simulation (LES) was performed using the standard Smagorinsky model to resolve highly unsteady flow features. Because the slope of CM with respect to AoA was negative for all cases, it was found that the vehicle is statically stable. For M0.9, M0.6 and M0.3 cases, absolute values of CM were almost the same. On the other hand, for M0.1, CM had a particularly large value, because the surface pressure distribution on rear side of the vehicle was different from the other cases. This difference was attributed to the separation point on the lower torus moving backward and turbulence in wake being enhanced with a decrease in Mach number and an increase in the Reynolds number.
  • Terashima, Hiroshi, Matsugi, Akira, Koshi, Mitsuo
    COMBUSTION AND FLAME 184 324 - 334 2017年10月 [査読有り][通常論文]
     
    The present study deals with the mechanisms for the hot-spot formation and pressure wave development associated with end-gas autoignition during knocking combustion of n-heptane/air mixture. The discussion is based on a one-dimensional (1-D) direct numerical simulation, where the compressible Navier-Stokes equations are solved with a detailed chemical kinetic mechanism of n-heptane, involving 373 species and 1071 reactions. The result demonstrates that the first trigger for a hot-spot formation is a compression wave generated by forced autoignition of a hot kernel and its reflection at a wall. The wall reflection of the propagating compression wave periodically produces an instantaneous temperature increase, which leads to the production of a larger amount of chemical species compared to that of other end-gas points. This non-uniform progress of chemical reaction process continues to exist at the wall, although the temperature increase is transient, resulting in faster autoignition and pressure wave generation at the wall. Thus, an important aspect of observing chemistry behaviors rather than temperature is demonstrated on the mechanism of hot-spot formation. The present study further addresses the reactivity of hot-spots on the relevance to pressure wave developments, wherein a zero-dimensional (0-D) ignition problem with pulsed compression waves is introduced. The higher reactivity of n-heptane/air mixture against pulse waves is observed with the faster ignition delay times in lower and higher initial temperature conditions. Conversely, the result at initial temperatures of 750-800 K indicates the lower reactivity with no significant effects of pulse waves on the ignition delay times. This is connected with the fuel characteristics of a negative temperature coefficient. Thus, in the 1-D simulations, a hot-spot with the high reactivity enhances spatial temperature difference in the end-gas region, leading to strong pressure wave generations. In contrast, a hot-spot with the low reactivity suppresses the pressure wave development with little spatial variation in temperature. The result demonstrates a significant aspect of hot-spot formation and reactivity on pressure wave development during knocking combustion. (C) 2017 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
  • 寺島洋史
    日本燃焼学会誌 59 189 149‐150 - 150 一般社団法人 日本燃焼学会 2017年08月 [査読無し][通常論文]
  • Akira Matsugi, Hiroshi Terashima, Hiroshi Terashima
    Combustion and Flame 179 238 - 241 2017年05月 [査読有り][通常論文]
     
    © 2017 The Combustion Institute The diffusive-thermal effect plays an important role in the intrinsic instability of premixed flames. A two-dimensional direct numerical simulation of the propagation of premixed hydrogen–air flames was performed using a detailed chemical kinetics model. The cellular behavior of a lean hydrogen–air flame was analyzed on the basis of its chemical structure. The primary consequence of the diffusive-thermal effect was found to be a change in the buildup process of reactive intermediates by the chain reaction mechanism at the preheat zone. The resultant chemical structure at the main reaction zone can be explained by the local composition change of the gas flowing into the reaction zone from the preheat zone.
  • Youhi Morii, Hiroshi Terashima, Mitsuo Koshi, Taro Shimizu, Eiji Shima
    JOURNAL OF COMPUTATIONAL PHYSICS 322 547 - 558 2016年10月 [査読有り][通常論文]
     
    We herein propose a fast and robust Jacobian-free time integration method named as the extended robustness-enhanced numerical algorithm (ERENA) to treat the stiff ordinary differential equations (ODEs) of chemical kinetics. The formulation of ERENA is based on an exact solution of a quasi-steady-state approximation that is optimized to preserve the mass conservation law through use of a Lagrange multiplier method. ERENA exhibits higher accuracy and faster performance in homogeneous ignition simulations compared to existing popular explicit and implicit methods for stiff ODEs such as VODE, MTS, and CHEMEQ2. We investigate the effects of user-specified threshold values in ERENA, to provide trade-off information between the accuracy and the computational cost. (C) 2016 Elsevier Inc. All rights reserved.
  • MUTO Daiki, TERASHIMA Hiroshi, TSUBOI Nobuyuki
    TRANSACTIONS OF THE JAPAN SOCIETY FOR AERONAUTICAL AND SPACE SCIENCES, AEROSPACE TECHNOLOGY JAPAN 14 30 Pa_45 - Pa_52 一般社団法人 日本航空宇宙学会 2016年 [査読有り]
     

    Effects of injector geometries on cryogenic co-flowing planar jets under a supercritical pressure are numerically investigated. The present study covers a wide range of injector exit geometries which focuses post lip height and recess length, and evaluates these effects on mixing characteristics. A hybrid ILES/RANS methodology is applied to simulate wall-bounded injector regions. The results show that thicker post lips generate larger vortices behind the post lip, resulting the comb-like structure of the rolled-up inner dense jet. As a result, the mixing is well improved, and the inner jet potential core is shortened. The recessed injectors additionally induce a large-scale flapping motion of the inner jet and further enhance the mixing. The frequency analysis with velocity fluctuations demonstrates that the vortex shedding behind the post lip has a frequency which depends on the post lip height. The recessed injectors induce another low-frequency peak of the flapping motion, and which value is independent of the post lip height.

  • Hiroumi Tani, Yutaka Umemura, Yu Daimon, Hiroshi Terashima, Mitsuo Koshi
    54th AIAA Aerospace Sciences Meeting 2016年 
    The vaporization and burning of the N2H4 and NTO droplets were simulated with the interface-tracking method to accurately explore the auto-ignition processes and the flame structures. The N2H4 vapor plume developed behind the N2H4 droplet and reacted with the ambient NO2 gas through the hydrogen abstraction reactions. Thus, the N2H4 vapor and NO2 gas mixtures behind the droplet were preheated and reached the auto-ignition at a few ms. The auto-ignition occurred in the multiple points almost at the same time. After the ignition, the premixed flame developed around the droplet. Thus, the vaporization of the liquid N2H4 near the surface became significant. Then, the double flame structures which comprise the inner decomposition flame and oxidation flame appeared around the droplet. The NTO droplet was not auto-ignited in the computational time of the present study because little N2O4 vapor near a saturated temperature decomposed to NO2 gas which is necessary for the hydrogen abstraction reactions. When the ignition was forced, the double flames developed. The outer decomposition flame propagated to the boundaries of the computational domain, while the inner oxidation flame appeared near the droplet. Except for the propagation of the decomposition flame, the NTO droplet combustion was similar to that of the industrial fuels.
  • Yu Daimon, Hiroumi Tani, Hiroshi Terashima, Mitsuo Koshi
    54th AIAA Aerospace Sciences Meeting 2016年 
    Hydrazine (N2H4)/nitrogen dioxide (NO2) un-like doublet impinging gas jets were simulated to explore the hypergolic ignition processes in a N2H4/N2O4 bipropellant thruster. The three-dimensional compressible Navier-Stokes equations with a detailed chemical kinetics mechanism, in which more than 200 chemical reactions were directly taken into account, were solved to reveal the influence of the chemical reaction. The differences of three-dimensional structures of hypergolic ignition process and mechanism of flame holding between the two different inlet gas temperatures of 400 and 600 K were discussed in order to investigate the influence of the induction time of chemical reaction on the three-dimensional flowfield. The computed results clarified that the ignition time of impinging gas jets can be significantly influenced by the ignition delay of the detailed chemical kinetics mechanism. In addition, the intermittent multi-ignitions played a significant role in the mechanism of flame holding.
  • Daiki Muto, Hiroshi Terashima, Nobuyuki Tsuboi
    54th AIAA Aerospace Sciences Meeting 2016年 
    Three-dimensional numerical simulations of cryogenic coaxial jets under supercritical pressure are performed with flushed and recessed injectors to investigate the effect of a recess on the coaxial mixing. A hybrid ILES/RANS method is applied to simulate wallbounded injectors and a recessed region. The recessed injector enhances the density decay and the temperature increase on the central axis, indicating the improvement of mixing compared with the flushed injector. However, the mixing improvement by the recess is not significant in the present conditions. The recess also induces distinct vortex rings around the outer jet. The power spectra of the velocity fluctuations also demonstrated that the low-frequency velocity fluctuations are clearly induced by the recess which frequency corresponds to the large vortex structures.
  • H. Terashima, Y. Daimon
    52nd AIAA/SAE/ASEE Joint Propulsion Conference, 2016 2016年 [査読無し]
     
    A two-dimensional detailed numerical simulation is performed for combustion flow field of a GCH4/GOX single injector using detailed chemical kinetics with the compressible Navier-Stokes equations. A detailed mechanism of CH4, 33 chemical species and 150 re- actions, is efficiently and directly introduced. The result shows that the relatively high- temperature and CH4-rich recirculation region is established in the upper and lower corners of the combustion chamber. The result, with a at inlet profile, interestingly shows the generation of an unstable combustion mode, which is not observed with a smooth inlet pro- file. It is shown that the disappearance of non-premixed flames behind the GOX post is a trigger for the unstable combustion mode through the production of partly premixed gases and the generation of autoignition at several locations in the combustion chamber, which may be caused by the extent of the incursion of GCH4 and GOX jets in the recirculation region behind the GOX post.
  • Soshi Kawai, Hiroshi Terashima, Hideyo Negishi
    JOURNAL OF COMPUTATIONAL PHYSICS 300 116 - 135 2015年11月 [査読有り][通常論文]
     
    This paper addresses issues in high-fidelity numerical simulations of transcritical turbulent flows at supercritical pressure. The proposed strategy builds on a tabulated look-up table method based on REFPROP database for an accurate estimation of non-linear behaviors of thermodynamic and fluid transport properties at the transcritical conditions. Based on the look-up table method we propose a numerical method that satisfies high-order spatial accuracy, spurious-oscillation-free property, and capability of capturing the abrupt variation in thermodynamic properties across the transcritical contact surface. The method introduces artificial mass diffusivity to the continuity and momentum equations in a physically-consistent manner in order to capture the steep transcritical thermodynamic variations robustly while maintaining spurious-oscillation-free property in the velocity field. The pressure evolution equation is derived from the full compressible Navier-Stokes equations and solved instead of solving the total energy equation to achieve the spurious pressure oscillation free property with an arbitrary equation of state including the present look-up table method. Flow problems with and without physical diffusion are employed for the numerical tests to validate the robustness, accuracy, and consistency of the proposed approach. (C) 2015 Elsevier Inc. All rights reserved.
  • 井上智博, 渡辺紀徳, 姫野武洋, 越光男, 寺島洋史
    微粒化 24 82 61 - 67 日本液体微粒化学会 2015年07月 [査読有り][通常論文]
  • Nozomu Kanno, Hiroumi Tani, Yu Daimon, Hiroshi Terashima, Norihiko Yoshikawa, Mitsuo Koshi
    JOURNAL OF PHYSICAL CHEMISTRY A 119 28 7659 - 7667 2015年07月 [査読有り][通常論文]
     
    The reactions of NO2 with cis-/trans-CH3NHNH, CH3NNH2 and CH2NHNH2 have been studied theoretically by quantum chemical calculations and steady-state unimolecular master equation analysis based on RRKM theory. The barrier heights for the roaming transition states between nitro (RNO2) and nitrite (RONO) isomerization reactions and those for the concerted HONO and HNO2 elimination reactions from RNO2 and RONO, affect the pressure dependences of the product-specific rate coefficients. At ambient temperature and pressure, the dominant product of the reactions of NO2 with cis-/trans-CH3NHNH and CH2NHNH2 would be expected to be HONO with trans-CH3NNH and CH2NNH2, respectively, whereas it is CH3N(NH2)NO2 for CH3NNH2 + NO2. The product-specific rate coefficients for the titled and related reactions on the same potential energy surfaces were proposed for kinetics modeling.
  • Hiroshi Terashima, Mitsuo Koshi
    COMBUSTION AND FLAME 162 5 1944 - 1956 2015年05月 [査読有り][通常論文]
     
    A knocking combustion modeled using a one-dimensional constant volume reactor is simulated in a manner of direct numerical simulations, in which large detailed chemical kinetic mechanisms for two pre-mixed gases, n-butane (113 species) and n-heptane (373 species), are directly and efficiently introduced. Detailed mechanisms of strong pressure wave generation during end-gas autoignition are clarified. Comparison of n-butane and n-heptane shows that the presence of the negative temperature coefficient (NTC) region significantly influences not only the timing of knocking occurrence but also the amplitude of pressure oscillations. In the case of n-heptane with the condition of an adiabatic wall, there is one large peak produced in the strength of knocking intensity for initial temperature between 450 and 1000 K, whereas there is no peak produced in the case of n-butane. The peak generated around 650 K is attributed to a pressure wave intensified through propagation in the end-gas, which is locally generated near the wall with the influence of the NTC region. It is also found that there is a transition of the autoignition position in the end-gas region from the wall for higher initial temperatures to the region ahead of the flame front for lower initial temperatures, leading to different mechanisms of knocking intensity generation. In the case of the isothermal wall condition, the peak around 650 K is reduced due to the lack of a local temperature increase at the wall, demonstrating the influence of wall temperature conditions on the strength of knocking intensity. (C) 2014 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
  • Youhi Morii, Hiroshi Terashima, Mitsuo Koshi, Taro Shimizu
    JOURNAL OF LOSS PREVENTION IN THE PROCESS INDUSTRIES 34 92 - 99 2015年03月 [査読有り][通常論文]
     
    A numerical simulation of the spontaneous ignition of high-pressure hydrogen in a duct with two obstacles on the walls is conducted to explore the spontaneous ignition mechanisms. Two-dimensional rectangular ducts are adopted, and the Navier Stokes equations with a detailed chemical kinetic mechanism are solved by using direct numerical simulations. In this study, we focus on the effects of the initial pressure of hydrogen and the position of the obstacles on the ignition mechanisms. Our results demonstrate that the presence of obstacles significantly changes the spontaneous ignition mechanisms producing three distinct ignition mechanisms. In addition, the position of the obstacles drastically changes the interaction of shock waves with the contact surface, and spontaneous ignition may take place at a relatively low pressure in some obstacle positions, which is attributed to the propagation direction and interaction timing of two reflected shock waves. (C) 2015 Elsevier Ltd. All rights reserved.
  • 谷 洋海, 寺島 洋史, 黒瀬 良一
    微粒化シンポジウム講演論文集 24 29 - 33 日本液体微粒化学会 2015年01月 [査読無し][通常論文]
  • Hiroumi Tani, Hiroshi Terashima, Yu Daimon, Mitsuo Koshi
    International Journal of Energetic Materials and Chemical Propulsion 14 1 71 - 84 2015年 [査読有り][通常論文]
     
    Hydrazine (N2H4)/nitrogen tetroxide (NTO) co-flowing jets were simulated to explore the hypergolic ignition processes in N2H4/NTO bipropellant thrusters. The Navier–Stokes equations with a detailed chemical kinetics mechanism were solved by direct numerical simulation to reveal the influence of the distinct chemical reaction i.e., hydrogen abstraction by nitrogen dioxide (NO2). In the ignition processes, the hydrogen abstraction sequence, especially the reaction of N2H3 and NO2, played a significant role in preheating the mixture gases and it was significant in the NTO-rich conditions rather than in the specifically stoichiometric conditions. Thus, in the N2H4/NTO co-flowing jets, the preheated regions correspond to the regions where the NTO-rich flows existed and contacted the N2H4 jet. The ignition eventually occurred in the region where NTO has sufficient concentration. Hence, it was found that the ignition timing and position strongly depend on whether the unsteady behavior of the NTO jet provides appropriate environments for the hydrogen abstraction reactions.
  • Hiroumi Tani, Hiroshi Terashima, Mitsuo Koshi, Yu Daimon
    PROCEEDINGS OF THE COMBUSTION INSTITUTE 35 2199 - 2206 2015年 [査読有り][通常論文]
     
    Hydrazine (N2H4)/nitrogen tetroxide (NTO) co-flowing plane jets were simulated to explore the hypergolic ignition processes and flame structures in N2H4/NTO bipropellant thrusters. The Navier-Stokes equations with a detailed chemical kinetics mechanism were solved in a manner of direct numerical simulation to reveal the influence of the distinct chemical reaction, i.e., hydrogen abstraction by nitrogen dioxide (NO2) and the thermal decomposition of N2H4. In the ignition processes, the hydrogen abstraction sequence played a significant role in preheating the mixture gases. Further, the ignition eventually occurred in the region where both N2H4 and NTO were sufficiently supplied for preheating. Hence, the ignition position and delay strongly depended on the fluid-mixing conditions. After the flames reached a steady state, the combustion flames uniquely comprised two types of flames, the outer diffusion flame and the inner decomposition flame. The outer diffusion flame came from the oxidization by NTO. The inner decomposition flame was caused and maintained by the heat transfer from the outer diffusion flame and a high rate of heat release from the thermal decomposition of N2H4. Because of the decomposition flame, the decomposition products such as NH3 and H-2 were the major constituents of the downstream combustion gases. (C) 2014 The Combustion Institute. Published by Elsevier Inc. All rights reserved.
  • Hiroshi Terashima, Youhi Morii, Mitsuo Koshi
    International Journal of Energetic Materials and Chemical Propulsion 14 3 177 - 196 2015年 [査読有り][通常論文]
     
    © 2015 by Begell House, Inc. A robust explicit time integration method for stiff chemical reaction equations is proposed and applied to zero-dimensional ignition and one-dimensional combustion flow problems. The proposed method based on a multi-time scale method significantly improves the robustness of the original method by introducing two new strategies: automatic adjustment of time step size for each characteristic group using a quasi-steady-state assumption and automatic reset of base time step size using two appropriate criteria. The results for several zero-dimensional ignition problems demonstrate the robustness and accuracy of the proposed method compared to existing explicit and implicit integration methods. The present study also provides a computational cost estimation of various terms in the governing equations using a one-dimensional combustion problem (knocking simulation), where the Navier– Stokes equations are coupled with the chemical reaction equations. As well as the zero-dimensional problems, the robustness and capability of the proposed method are demonstrated. While the proposed method alleviates the occupancy of chemical reaction part in the total computational cost compared to an implicit time integration method, it is found that the transport properties calculations relatively increase with considerable amounts, suggesting efficient modeling of transport properties calculations for multi-dimensional combustion problems.
  • Akira Kikusato, Kazuya Kogo, Beini Zhou, Kusaka Jin, Yasuhiro Daisho, Kiyotaka Sato, Hidefumi Fujimoto, Hiroshi Terashima, Youhi Morii
    SAE Technical Papers 2014-October 2014年10月13日 [査読有り][通常論文]
     
    Copyright © 2014 SAE International. The objective of the present study is to analyze soot formation in diesel engine combustion by using multi-dimensional combustion simulations with a parallelized explicit ODE solver. Parallelized CHEMEQ2 was used to perform detailed chemical kinetics in KIVA-4 code. CHEMEQ2 is an explicit stiff ODE solver developed by Mott et al. which is known to be faster than traditional implicit ODE solvers, e.g., DVODE. In the present study, about eight times faster computation was achieved with CHEMEQ2 compared to DVODE when using a single thread. Further, by parallelizing CHEMEQ2 using OpenMP, the simulations could be run not only on calculation servers but also on desktop machines. The computation time decreases with the number of threads used. The parallelized CHEMEQ2 enabled combustion and emission characteristics, including detailed soot formation processes, to be predicted using KIVA-4 code with detailed chemical kinetics without the need for reducing the reaction mechanism. After validating the code, diesel engine combustion was simulated to investigate combustion and emission characteristics, focusing on soot formation, growth and oxidation at different EGR ratios. To predict soot formation, a gas-phase polycyclic aromatic hydrocarbons (PAH) precursor formation model was coupled with a detailed phenomenological particle formation model, which included soot nucleation from precursors, surface growth/oxidation and particle coagulation. The results indicate that increased soot emission at high EGR ratios is mainly caused by decreased oxidation by oxygen and OH radicals because mixing fuel and gases (including oxygen and OH) has significant effects on reducing the mass of soot.
  • Nozomu Kanno, Hiroshi Terashima, Yu Daimon, Norihiko Yoshikawa, Mitsuo Koshi
    INTERNATIONAL JOURNAL OF CHEMICAL KINETICS 46 8 489 - 499 2014年08月 [査読有り][通常論文]
     
    The kinetics and mechanisms of H atom abstraction reactions from CH3NHNH2 by NO2 (R1) and related reactions have been investigated theoretically by using B97X-D and CCSD(T)-F12 quantum chemical calculations and the steady-state unimolecular master equation analysis based on Rice-Ramsperger-Kassel-Marcus (RRKM) theory. For reaction (R1), both dissociation and isomerization steps between intermediate complexes were found to be important for the distribution of the dissociated bimolecular products. The dominant products of (R1) were found to be cis-CH3NHNH and HONO at lower temperature. The branching ratios for CH3NNH2 formation paths increased with increasing temperature. On the same reaction potential energy surface, six reactions including isomerization reactions between CH3NNH2 and cis-/trans-CH3NHNH catalyzed by HONO were suggested to compete with the reverse reaction of (R1). The temperature- and pressure-dependent rate expressions are proposed for kinetic modeling.
  • 森井雄飛, 寺島洋史, 越光男, 清水太郎
    日本燃焼学会誌 56 176 156 - 165 日本燃焼学会 2014年05月 [査読有り][通常論文]
  • Yu Daimon, Hiroshi Terashima, Mitsuo Koshi
    JOURNAL OF PROPULSION AND POWER 30 3 707 - 716 2014年05月 [査読有り][通常論文]
     
    A detailed chemical kinetic mechanism for hypergolic ignition of N2H4 in a N2O4-NO2 gas mixture has been constructed. In this mechanism, the hypergolic ignition is mainly caused by the sequential reactions of H atom abstraction from N2Hm by NO2: N2Hm + NO2 = N2Hm-1 HONO/HNO2 (m = 4 similar to 1). Although the first step of the H atom abstraction (m = 4) is endothermic, consecutive abstraction reactions for m = 3, 2, and I are exothermic, and especially heat release by the reaction of N2H + NO2 = N-2 + HONO(m =1) is large because of N-2 production. Temperature rise caused by the heat release accelerates the endothermic initiation reaction (m = 4). This thermal feedback is responsible for the hypergolic ignition at low temperatures. Because no experimental and theoretical information is available on these reactions, rate coefficients were evaluated on the basis of transition state theory, unimolecular rate theory, and master equation analysis with quantum chemical calculations of potential energy curves. In addition, reactions of N2H4 with N2O4 isomers were also examined. The present kinetic mechanism can explain gas-phase hypergolic ignition of N2H4/NTO mixtures at temperatures down to 200 K.
  • Hiroshi Terashima, Mitsuo Koshi, Chika Miwada, Toshio Mogi, Ritsu Dobashi
    INTERNATIONAL JOURNAL OF HYDROGEN ENERGY 39 11 6013 - 6023 2014年04月 [査読有り][通常論文]
     
    A two-dimensional (2-D) simulation of spontaneous ignition of high-pressure hydrogen in a length of duct is conducted to explore ignition mechanisms. The present study adopts a 2-D rectangular duct and focuses on effects of the initial diaphragm shape on spontaneous ignition. The Navier-Stokes equations with a detailed chemical kinetics mechanism are solved in a manner of direct numerical simulation. The detailed mechanisms of spontaneous ignitions are discussed for each initial diaphragm shape. For a straight diaphragm, ignition only occurs near the wall owing to the adiabatic wall condition, while three ignition events are identified for a greatly deformed diaphragm: ignition due to reflection of leading shock wave at the wall, hydrogen penetration into shock-heated air near the wall, and deep penetration of hydrogen into shock-heated air behind the leading shock wave. Copyright (C) 2014, Hydrogen Energy Publications, LLC. Published by Elsevier Ltd. All rights reserved.
  • Numerical study on mixing characteristics of coaxial cryogenic N2/H2 injection under supercritical pressure
    Hiroshi Terashima, Mitsuo Koshi
    52nd Aerospace Sciences Meeting 2014年 
    Three-dimensional simulations for nitrogen/hydrogen mixing of a shear coaxial injector under supercritical pressures are conducted in order to investigate their mixing characteristics. A high-order numerical method based on a sixth-order compact scheme is applied. The present study parametrically covers two pressures of 4 and 8 MPa and two outer hydrogen jet conditions, while an inner nitrogen jet condition and a mixture ratio are kept to be same. The discussions on mean and fluctuation properties in addition to instantaneous flow fields are made in order to characterize mixing features for the coaxial injection. A colder outer hydrogen jet provides less mixing behavior compared to a warmer hydrogen jet, generating a longer dense-core of the inner nitrogen jet unmixed nitrogen and hydrogen remain in downstream regions. Some unique features are found in the temperature profiles on the centerline of inner nitrogen jet as a flattened or a wave-like profile under 4 MPa, while no such unique features are not observed under 8 MPa. In the case of a warmer outer hydrogen jet, the hydrogen density show a maximum peak just after the injection because of the heat transfer to the inner nitrogen jet. The density and temperature fluctuation distributions are determined by the unique thermodynamic fluid variations around the critical temperature, and the velocity and temperature differences among two jets and chamber fluid. Finally, a clear dependence of dense-core length on the momentum flux ratio is demonstrated on comparison with experimental and computational data.
  • ISHIKAWA Katsutoshi, UMEMURA Yutaka, HIMENO Takehiro, WATANABE Toshinori, TANI Naoki, TERASHIMA Hiroshi, KOSHI Mitsuo
    TRANSACTIONS OF THE JAPAN SOCIETY FOR AERONAUTICAL AND SPACE SCIENCES, AEROSPACE TECHNOLOGY JAPAN 12 29 Pa_63 - Pa_69 一般社団法人 日本航空宇宙学会 2014年 [査読有り]
     
    Cavitation may cause a severe damage to rocket engine turbo-pump, and appropriate handling of cavitation is one of the key technologies for rocket engine development. Numerical simulation is quite useful technique, however, simulation result strongly relies on cavitation model. Cavitation flow is essentially strong unsteady and multi scale phenomena, scales of bubbles vary in size from small to large. Therefore, prediction of cavitation by numerical simulation is difficult. Many cavitation models have been proposed and researched. However, direct interface tracking approach has not been applied to cavitating flow and the model characteristic is not fully understood. This research shows direct interface tracking method can properly capture both cloud cavitation and super cavitation and grid resolution affects lift coefficient.
  • Youhi Morii, Hiroshi Terashima, Mitsuo Koshi, Taro Shimizu, Eiji Shima
    50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference 2014 2014年 [査読無し][通常論文]
     
    © 2014 by authors. Published by the American Institute of Aeronautics and Astronautics, Inc. A simple yet robust and fast time integration method is proposed for efficiently solving stiff chemical kinetic ordinary differential equations. The proposed method is based on a general formula which preserves the conservation laws for any integration operators con- structed using the Lagrange multiplier method. A quasi-steady-state approxixmation is used as the integrator. The time step size is automatically controlled by using a Lagrange multiplier so that the error, which is caused by the Lagrange multiplier method, is small. The results of several ignition problems demonstrate the robustness and accuracy of the proposed method in comparison with other integration methods such as a implicit inte- gration method (VODE), a multi time-scalse method (MTS), and a modified CHEMEQ2. The proposed method, named ERENA, provides the fastest performance for the most of conditions used in this study.
  • Daiki Muto, Hiroshi Terashima, Nobuyuki Tsuboi
    TRANSACTIONS OF THE JAPAN SOCIETY FOR AERONAUTICAL AND SPACE SCIENCES, AEROSPACE TECHNOLOGY JAPAN 12 ists29 Po_2_39-Po_2_44 - Po_2_44 THE JAPAN SOCIETY FOR AERONAUTICAL AND SPACE SCIENCES 2014年 [査読有り][通常論文]
     
    Investigations of real gas effects of supercritical fluids on shock tube problems were numerically performed. Understandings of fluid behaviors under supercritical conditions are essential for the design of high pressure combustors. We developed a numerical code for simulations of flow fields under the supercritical conditions. Cubic types of equation of state are applied to evaluate properties of nearcritical and supercritical fluids. The present code was tested on shock tube problems, and the present result agreed well with a reference result. Comparisons of results between the real gas equations of state and the ideal gas equation of state showed differences in position of shock waves, contact discontinuities, and expansion fans due to the specific thermodynamic features of the supercritical fluids. In a condition that crosses the critical point, the results showed a large density jump with a small temperature jump at the contact discontinuity because of drastic changes of thermodynamic properties. This indicated that the real gas effects obviously appear when the initial condition is close to the critical point of the working fluids.
  • Hiroshi Terashima, Soshi Kawai, Mitsuo Koshi
    Computers and Fluids 88 484 - 495 2013年12月15日 [査読有り][通常論文]
     
    An interface-capturing method using a high-order central difference scheme is presented for simulations of compressible multicomponent flows. The present method adds consistent numerical diffusion terms to robustly capture interface discontinuities, while maintaining the velocity, pressure, and temperature equilibriums at interfaces. Analysis of the numerical errors generated at the interfaces leads to a proposal of new consistent numerical diffusion terms. The method solves a fully conservative form of the total mass, momentum, total energy, and species-mass, and an additional advection form of the specific heats ratio to preserve the pressure oscillation-free property at the interfaces. Several one- and two-dimensional problems are used to verify the proposed method. © 2013 Elsevier Ltd.
  • 寺島洋史, 越光男
    日本燃焼学会誌 55 174 411 - 421 日本燃焼学会 2013年11月 [査読有り][通常論文]
  • Hiroshi Terashima, Mitsuo Koshi
    Journal of Propulsion and Power 29 6 1328 - 1336 2013年11月 [査読有り][通常論文]
     
    Two-dimensional planar and three-dimensional round nitrogen jets under supercritical pressures are simulated with a wide range of conditions to explore their unique characteristics. A high-order method using a sixth-order compact scheme is applied. The present study parametrically covers two supercritical pressures of 4 and 8 MPa and three jet injection temperatures between a cryogenic jet of 82Kand a warmer jet of 133 K. Unique characteristics are found in both the mean temperature and temperature fluctuation profiles on the centerline (i.e., slower increase of jet temperature and relatively weak temperature fluctuation), only in the case of a near-critical pressure of 4 MPa and a cryogenic jet of 82 K. For the condition, the temperature profile on the centerline uniquely consists of four characteristic regions. The other conditions show no distinct features. The present study suggests that the distributions of the specific heat capacity at constant pressure help to explain the generation of the unique temperature characteristics. Copyright © 2013 by the von Karman Institute for Fluid.
  • Chihiro Inoue, Mitsuo Koshi, Hiroshi Terashima, Takehiro Himeno, Toshinori Watanabe
    Science and Technology of Energetic Materials 74 3-4 106 - 111 2013年11月 [査読有り][通常論文]
     
    The physics behind the beauty of sparkling fireworks has not been clarified yet due to a lack of detailed visualization results. In the present study, atomization process in sparkling fireworks is elucidated by using a high-speed video camera. In the first-half sequence of the fireworks, the fireball repeatedly expands, bursts, and shrinks due to the high pressure gas inside the fireball. In contrast, in the last-half sequence, the bubbly fireball slightly deforms, and small bubbles burst on the fireball. A scenario of droplets generation is as follows : a liquid thread extends from the bottom of the bursting fireball, and fragments into droplets. Thus the droplets originate from inside the fireball rather than from its surface.
  • Hiroshi Terashima, Mitsuo Koshi
    Computers and Fluids 85 39 - 46 2013年10月01日 [査読有り][通常論文]
     
    This study presents a strategy for simulations of cryogenic single-species jets under supercritical pressure conditions. In this strategy, a pressure evolution equation is introduced and numerical diffusion terms are consistently constructed in order to maintain the pressure and velocity equilibriums at fluid interfaces. By taking the idea of the equilibrium, the interfaces with high density and temperature ratio are robustly captured without the generation of spurious oscillations, while a high-order central differencing scheme resolves the flow fields. The present method preserves the mass and momentum conservation properties, while the poor energy conservation property is recognized. The one-dimensional interface advection and two-dimensional cryogenic jet mixing problems demonstrate the superiority and robustness of the present method over a conventional fully conservative approach. © 2012 Elsevier Ltd.
  • 井上 智博, 越 光男, 寺島 洋史, 姫野 武洋, 渡辺 紀徳
    Science and Technology of Energetic Materials : journal of the Japan Explosive Society 74 3 106 - 111 火薬学会 2013年08月 [査読有り][通常論文]
     
    The physics behind the beauty of sparkling fireworks has not been clarified yet due to a lack of detailed visualization results. In the present study, atomization process in sparkling fireworks is elucidated by using a high-speed video camera. In the first-half sequence of the fireworks, the fireball repeatedly expands, bursts, and shrinks due to the high pressure gas inside the fireball. In contrast, in the last-half sequence, the bubbly fireball slightly deforms, and small bubbles burst on the fireball. A scenario of droplets generation is as follows : a liquid thread extends from the bottom of the bursting fireball, and fragments into droplets. Thus the droplets originate from inside the fireball rather than from its surface.
  • Yu Daimon, Hiroshi Terashima, Mitsuo Koshi
    Science and Technology of Energetic Materials 74 1-2 17 - 22 2013年07月 [査読有り][通常論文]
     
    Hydrazine (N2H4) has a unique characteristic inducing hypergolic ignition even at very low temperatures with nitrogen dioxide (NO2). In order to understand the chemical kinetic origin of this hypergolic nature, thermochemical data (heat of formation, specific heat capacity, and entropy) for chemical species relevant to N2H 4/NO2 combustion are firstly evaluated on the basis of quantum chemical calculation at the CBS-QB3 level of theory. Then, a preliminary detailed chemical kinetic mechanism for gas-phase combustion of N 2H4/NO2 mixtures has been constructed. Kinetic simulations indicated that sequential reactions of N2H m+NO2 (m=4,3,2,l), that is, N2H 4+NO2=N2H3+HONO (1), N 2H3H-NO2=N2H3NO 2= N2H2H-HONO (2), N2H 2+NO2=NNH+HONO (3), and NNH+NO2=N 2+HONO (4), are responsible for the hypergolic ignition at low temperatures. Rate constants of these reactions were estimated based on the transition state theory and unimolecular rate theory. The proposed mechanism can predict low temperature ignition of N2H4/NO 2mixtures. The origin of the low temperature ignition is the reaction sequence of hydrogen abstraction by NO2 from N2H 4, N2H4=>N2H3= >N2H2=>NNH=>N2, Large amount of heat is released during this reaction sequence, especially by the reaction (4) which produces N2, and the resulting temperature rise accelerates the reaction (1), which has a small activation barrier.
  • Hiroshi Terashima, Mitsuo Koshi
    51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition 2013 2013年 
    Two-dimensional planar and three-dimensional round nitrogen jets under supercritical pressures are simulated with a wide range of conditions in order to explore its unique characteristics. A high-order method using a sixth-order compact scheme is applied. The present study parametrically covers two supercritical pressures of 4 MPa and 8 MPa and three jet injection temperatures between a cryogenic jet of 82 K and a warmer jet of 133 K. Unique characteristics are found in both the mean temperature and the temperature fluctuation profiles on the centerline, i.e., slower increase of jet temperature and relatively weak temperature fluctuation, only in case of a near-critical pressure of 4 MPa and a cryogenic jet of 82 K. The temperature profile on the centerline of the cryogenic jet of 82 K under 4 MPa uniquely consists of four characteristic regions. The other conditions show no distinct features. The present study suggests that the distributions of the specific heat capacity at constant pressure help to explain the generation of the unique characteristics on the temperature. © 2013 by the authors.
  • Yu Daimon, Hiroshi Terashima, Mitsuo Koshi
    51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition 2013 2013年 
    A detailed chemical kinetic mechanism for hypergolic ignition of N2H4/NO2 gas mixture at low temperatures has been constructed. In this mechanism, the hypergolic ignition is caused by following sequential reactions of H atom abstraction from N2Hm by NO2. N2H4+ NO2= N2H3+ HONO or HNO2(R1) N2H3+ NO2= N2H2+ HONO (R2) N2H2+ NO2= N2H + HONO (R3) N2H + NO2= N2+ HONO (R4) These reactions are exothermic, especially heat release by the reaction (R4) is large because of N2 production. Temperature rise caused by the heat release accelerates the initiation reaction (R1). This 'thermal feedback' is responsible to the hypergolic ignition at ambient temperatures. Since no experimental and theoretical information was available on these reactions, rate coefficients were evaluated on the basis of transition state theory, uni-molecular rate theory, and master equation analysis with quantum chemical calculations of potential energy curves. Results of simulations by using the present mechanism including reactions (R1)-(R4) reasonably agree with existing experimental data. © 2013 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.
  • Hiroshi Terashima, Mitsuo Koshi
    JOURNAL OF COMPUTATIONAL PHYSICS 231 20 6907 - 6923 2012年08月 [査読有り][通常論文]
     
    This study proposes an approach for simulations of cryogenic fluid mixing under supercritical pressures using high-order schemes. In this approach, we introduce a pressure evolution equation and consistently construct numerical diffusion terms to maintain the velocity and pressure equilibriums at fluid interfaces. The interfaces with high density and temperature ratio are successfully captured without the generation of spurious oscillations, while a high-order central differencing scheme resolves the flow fields. The present method preserves the mass and momentum conservation properties, while the poor energy conservation property is recognized. The one-dimensional single and multi-species interface advection and two-dimensional cryogenic jet mixing problems demonstrate the superiority and robustness of the present method over a conventional fully conservative method. (C) 2012 Elsevier Inc. All rights reserved.
  • Taku Nonomura, Seiichiro Morizawa, Hiroshi Terashima, Shigeru Obayashi, Kozo Fujii
    JOURNAL OF COMPUTATIONAL PHYSICS 231 8 3181 - 3210 2012年04月 [査読有り][通常論文]
     
    A weighted compact nonlinear scheme (WCNS) is applied to numerical simulations of compressible multicomponent flows, and four different implementations (fully or quasi-conservative forms and conservative or primitive variables interpolations) are examined in order to investigate numerical oscillation generated in each implementation. The results show that the different types of numerical oscillation in pressure field are generated when fully conservative form or interpolation of conservative variables is selected, while quasi-conservative form generally has poor mass conservation property. The WCNS implementation with quasi-conservative form and interpolation of primitive variables can suppress these oscillations similar to previous finite volume WENO scheme, despite the present scheme is finite difference formulation and computationally cheaper for multi-dimensional problems. Series of analysis conducted in this study show that the numerical oscillation due to fully conservative form is generated only in initial flow fields, while the numerical oscillation due to interpolation of conservative variables exists during the computations, which leads to significant spurious numerical oscillations near interfaces of different component of fluids. The error due to fully conservative form can be greatly reduced by smoothing interface, while the numerical oscillation due to interpolation of conservative variables cannot be significantly reduced. The primitive variable interpolation is, therefore, considered to be better choice for compressible multicomponent flows in the framework of WCNS. Meanwhile better choice of fully or quasi-conservative form depends on a situation because the error due to fully conservative form can be suppressed by smoothed interface and because quasi-conservative form eliminates all the numerical oscillation but has poor mass conservation. (C) 2012 Elsevier Inc. All rights reserved.
  • Hiroshi Terashima, Soshi Kawai, Mitsuo Koshi
    American Society of Mechanical Engineers, Fluids Engineering Division (Publication) FEDSM 1 635 - 643 2012年 [査読無し][通常論文]
     
    We present an interface-capturing method for fluid interfaces in compressible multicomponent flows using high-order central-difference-based schemes. Numerical diffusion terms are consistently designed so that the velocity, pressure, and temperature equilibriums are maintained at the fluid interfaces, while serving as an efficient interface-capturing. Advection problems of a contact discontinuity and a material interface shows that 1) the present method maintains the velocity, pressure, and temperature equilibriums at the fluid interfaces (oscillation-free property) and 2) the numerical diffusion terms effectively works for suppressing spurious wiggles of the density or temperature. Comparisons with a conventional fully-conservative approach demonstrates the superiority of the present method in avoiding spurious oscillations. A shock tube problem of two-component gases shows the capability for capturing the shock wave while the velocity and pressure equilibriums are successfully maintained at the contact discontinuity. Copyright © 2012 by ASME.
  • Hiroshi Terashima, Soshi Kawai, Nobuhiro Yamanishi
    AIAA JOURNAL 49 12 2658 - 2672 2011年12月 [査読有り][通常論文]
     
    A high-resolution methodology using a high-order compact differencing scheme with localized artificial diffusivity is introduced with the aim of simulating jet mixing under supercritical pressure environments. The nonlinear localized artificial diffusivity provides the stability to capture different types of discontinuity, such as shock wave, contact surface, and material interface, whereas the high-order compact difference scheme resolves broadband scales in the rest of the domain. The present method is tested on several one-dimensional discontinuity-related problems under super/transcritical conditions and a comparatively more illustrative two-dimensional low-temperature planar jet problem under a supercritical pressure condition. The localized artificial diffusivity, especially artificial thermal conductivity for temperature gradients, effectively suppresses numerical wiggles near the interfaces. The effects of the artificial thermal conductivity on numerical stability and accuracy are examined. Comparisons between the present method and a conventional low-order scheme demonstrate the superior performance of the present method for resolving a wide range of flow scales while successfully capturing large density/temperature variations at interfaces.
  • Soshi Kawai, Hiroshi Terashima
    INTERNATIONAL JOURNAL FOR NUMERICAL METHODS IN FLUIDS 66 10 1207 - 1225 2011年08月 [査読有り][通常論文]
     
    A simple methodology for a high-resolution scheme to be applied to compressible multicomponent flows with shock waves is investigated. The method is intended for use with direct numerical simulation or large eddy simulation of compressible multicomponent flows. The method dynamically adds non-linear artificial diffusivity locally in space to capture different types of discontinuities such as a shock wave, contact surface or material interface while a high-order compact differencing scheme resolves a broad range of scales in flows. The method is successfully applied to several one-dimensional and two-dimensional compressible multicomponent flow problems with shock waves. The results are in good agreement with experiments and earlier computations qualitatively and quantitatively. The method captures unsteady shock and material discontinuities without significant spurious oscillations if initial start-up errors are properly avoided. Comparisons between the present numerical scheme and high-order weighted essentially non-oscillatory (WENO) schemes illustrate the advantage of the present method for resolving a broad range of scales of turbulence while capturing shock waves and material interfaces. Also the present method is expected to require less computational cost than popular high-order upwind-biased schemes such as WENO schemes. The mass conservation for each species is satisfied due to the strong conservation form of governing equations employed in the method. Copyright (C) 2010 John Wiley & Sons, Ltd.
  • Hiroshi Terashima, Soshi Kawai, Nobuhiro Yamanishi
    41st AIAA Fluid Dynamics Conference and Exhibit 2011年 
    Two-dimensional planar nitrogen jets in supercritical thermodynamic conditions are simulated using a high-resolution numerical method (which consists of a sixth-order compact difference scheme and a localized artificial diffusivity method) with the aim at exploring its unique characteristics. Effects of injection temperature, chamber pressure, and the equation of state on supercritical jet behaviors are investigated. Throughout the investigations, two major unique characteristics are found under the transcritical conditions. One unique characteristic are found in the mean temperature profile, as the slower increase of jet temperature, in the case of transcritical injections. The cause for the unique feature is simply explained by the corresponding temperature-density (T-ρ) diagrams. Another unique feature of supercritical jet flows appears in the generation of different flow scales. The power spectrums and the flow fields quantitatively and qualitatively show the different features of flow scales due to the injection conditions. In the transcritical injection, the production of smaller flow scales is considerably enhanced relative to the other injection cases, due to not only its higher density ratio, but also its abrupt variations (leap) of δρ/δT in the T-ρ diagrams. This study indicates, for the present conditions used, that the unique characteristics of supercritical jet flows appear in the mean temperature distributions and the generation of different flow scales, which are simply yet effectively explained by the T-ρ diagrams. © 2011 by the authors.
  • Hiroshi Terashima, Gretar Tryggvason
    COMPUTERS & FLUIDS 39 10 1804 - 1814 2010年12月 [査読有り][通常論文]
     
    A front-tracking method for compressible multi-fluid flows is presented, where marker points are used both for tracking fluid interfaces and also for constructing the Riemann problem on the interfaces. The Riemann problem between the two fluid phases (defined in the interface normal direction) is solved using the exact Riemann solver on the marker points. The solutions are projected onto fixed grid points and then extrapolated into the corresponding ghost-fluid regions, to be used as boundary conditions. Each fluid phase is solved separately as in the ghost-fluid method. The proposed procedures, especially the projection of the exact Riemann solutions onto the fluid grids, are designed to be simple and consistent in any spatial dimensions. Several multi-fluid problems, including the breakup of a water cylinder induced by the passage of a shock wave were computed in order to demonstrate the capability of the proposed method. (C) 2010 Elsevier Ltd. All rights reserved.
  • Hiroshi Terashima, Gretar Tryggvason
    JOURNAL OF COMPUTATIONAL PHYSICS 228 11 4012 - 4037 2009年06月 [査読有り][通常論文]
     
    A front-tracking/ghost-fluid method is introduced for simulations of fluid interfaces in compressible flows. The new method captures fluid interfaces using explicit front-tracking and defines interface conditions with the ghost-fluid method. Several examples of multi-phase flow simulations, including a shock-bubble interaction. the Richtmyer-Meshkov instability, the Rayleigh-Taylor instability, the collapse of an air bubble in water and the breakup of a water drop in air, using the Euler or the Navier-Stokes equations, are performed in order to demonstrate the accuracy and capability of the new method. The computational results are compared with experiments and earlier computational studies. The results show that the new method can simulate interface dynamics accurately, including the effect of surface tension. Results for compressible gas-water systems show that the new method can be used for simulations of fluid interface with large density differences. (C) 2009 Elsevier Inc. All rights reserved.
  • Hiroshi Terashima, Kenji Ono
    Computational Fluid Dynamics 2006 - Proceedings of the Fourth International Conference on Computational Fluid Dynamics, ICCFD 2006 843 - 848 2009年
  • Hiroshi Terashima, Kenji Ono
    2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007 1 SYMPOSIA PART A 219 - 228 2007年 
    A compressible flow solver coupled with moving/deformed geometries on Cartesian grid with Signed Distance Field (SDF) is developed and its capability is investigated through computations of several basic flow fields for future applications with certain reliability. The flow solver is designed so that SDF includes sufficient geometrical information to compute flow fields. Since information of moving/deformed geometries is recognized as a change of the SDF between time steps, the flow solver can be coupled with moving/deformed geometries naturally. The implementation of this solver is simple and easy. No modification is needed in the main part of the flow solver. Furthermore, the interpolation and the corresponding stencils searching process are not required. Several basic flow fields around fixed/moving cylinders and a fixed sphere are computed in order to validate the proposed solver, in which the computed results are compared with available numerical and experimental results. The results demonstrated the method's capability for moderate Reynolds number flows around both of fixed and moving geometries. Based on the results, some criteria and problems for obtaining reliable solution are suggested. Copyright © 2007 by ASME.
  • Hiroshi Terashima, Kenji Ono
    FEDSM 2007: PROCEEDINGS OF THE 5TH JOINT AMSE/JSME FLUIDS ENGINEERING SUMMER CONFERENCE VOL 1, PTS A AND B 219 - 228 2007年 
    A compressible flow solver coupled with moving/deformed geometries on Cartesian grid with Signed Distance Field (SDF) is developed and its capability is investigated through computations of several basic flow fields for future applications with certain reliability. The flow solver is designed so that SDF includes sufficient geometrical information to compute flow fields. Since information of moving/deformed geometries is recognized as a change of the SDF between time steps, the flow solver can be coupled with moving/deformed geometries naturally. The implementation of this solver is simple and easy. No modification is needed in the main part of the flow solver. Furthermore, the interpolation and the corresponding stencils searching process are not required. Several basic flow fields around fixed/moving cylinders and a fixed sphere are computed in order to validate the proposed solver, in which the computed results are compared with available numerical and experimental results. The results demonstrated the method's capability for moderate Reynolds number flows around both of fixed and moving geometries. Based on the results, some criteria and problems for obtaining reliable solution are suggested.
  • Hiroshi Terashima, Kozo Fujii
    AIAA JOURNAL 45 1 237 - 246 2007年01月 [査読有り][通常論文]
     
    Transonic and supersonic flutter characteristics of a delta wing configuration with external stores were computationally simulated, and the aerodynamic influence of the stores on the flutter characteristics was investigated. Delta wings with one and two external stores were considered. Unsteady aerodynamics of the wing with external stores were evaluated using Navier-Stokes equations, and equations of motion based on a modal approach were applied to the structural dynamics. These equations were,coupled using a subiteration approach. The computational results showed that the flutter dynamic pressures were reduced for a wide range of Mach numbers when the external stores were attached. The flutter dynamic pressures also decreased as the number of external stores increased. In the case of one external store, the aerodynamic influence of the store could be divided into two regions according to the Mach number. It was found that neglecting the aerodynamic influence of the store led to an overestimation of the flutter dynamic pressures in the supersonic flow region. In the case of two external stores, the aerodynamic influence of the stores on the flutter boundary appeared only at one supersonic Mach number, unlike the case of one external store. Additional flutter analysis using a different store size and unsteady aerodynamic analysis with forced oscillations indicated that the interference position of the shock wave generated ahead of the external store on the lower surface was a key factor in determining the flutter boundary, and the shock wave oscillation may have acted as a negative damping on the wing motion.
  • 寺島 洋史, 藤井 孝藏
    日本機械学會論文集. B編 = Transactions of the Japan Society of Mechanical Engineers. B 71 712 2856 - 2863 一般社団法人日本機械学会 2005年 [査読有り][通常論文]
     
    One criterion that determines time stepping size in the implicit time integration method is given for accurate and effective transonic flutter simulations. Transonic flows over two-dimensional forced oscillating air foils with several reduced frequencies are first investigated for constructing a criterion for selecting the time stepping size, and then two and three-dimensional transonic flutter simulations are performed for evaluating the criterion. Results for the forced oscillating airfoil indicate that unsteady aerodynamic forces converge in a constant value as the time stepping size decreases and the time stepping size required the convergence of unsteady aerodynamic force depends on the reduced frequency. From these results, it turns out that 4000 integration steps during one cycle of airfoil oscillations are enough for the estimation of the unsteady aerodynamic forces at any reduced frequency. The time stepping size is automatically decided by setting 4000 integration steps during one cycle of airfoil oscillations. Results for two and three-dimensional transonic flutter simulations show that flutter boundaries can be accurately and effectively calculated based on the criterion.
  • Hiroshi Terashima, Kozo Fujii
    34th AIAA Fluid Dynamics Conference and Exhibit 2004年 
    Transonic and supersonic flutter characteristics of a delta wing configuration with external stores are simulated using the fluid/structure coupling method and the effect of the store aerodynamics is investigated. A delta wing with one and two external stores are considered in this study. The stores are located in the rearward of the wing simulating typical high speed aircraft configurations. The computational result shows that the flutter dynamic pressures over all the Mach numbers are reduced by adding external stores for both cases. In addition, the flutter dynamic pressures become much lower when the number of attached external stores is increased. In the case of the single external store, influence of the store aerodynamics can be divided into two regions according to the Mach number. It is found that neglecting the store aerodynamics leads to the overestimation of the flutter dynamic pressures in the supersonic region. In case of two external stores, the store aerodynamics influence on the flutter boundary only at M∞ = 1.22 unlike the case of the single external store. It is shown that the shock wave ahead of the external store influences on the pressure distributions on the lower surface and the location of the shock wave is important for the decision of the flutter boundary. © 2004 by the American Institute of Aeronautics and Astronautics, Inc.
  • 寺島 洋史, 藤井 孝藏
    年次大会講演論文集 2002 293 - 294 一般社団法人日本機械学会 2002年 
    Flutter phenomena for the AGARD445.6 standard aeroelastic wing in the transonic flow are simulated using the fluid/structure coupling method. The effect of time accuracy, number of the inner-iterations and comparison of a fully implicit coupling method with a loosely coupling method are examined. The sufficient number of inner-iterations in time is required for the accurate prediction of time responses when using a large time step. Improvement of the solution time accuracy for the aerodynamic equations is key for accurate aeroelastic computations and the one-step time lag generated in a loosely coupling method does not have an effect in the flutter prediction in transonic flows.

MISC

書籍等出版物

  • Bellan, Josette, American Institute of Aeronautics and Astronautics (範囲:Detailed Modeling of Supercritical Jets and Flames, pp. 571–630)
    American Institute of Aeronautics and Astronautics 2020年 (ISBN: 9781624105807) xiv, 787 p.
  • 1) 計算力学ハンドブック「第II巻 差分法・有限体積法(熱流体編)」
    寺島洋史 (担当:分担執筆範囲:第4章圧縮性流れ第5項連成問題)
    日本機械学会,丸善 2006年

講演・口頭発表等

  • Shun Murakami, Hiroshi Terashima, Nobuyuki Oshima
    AIAA Scitech 2019 Forum 2019年01月 
    © 2019 by Timothy K. Minton. Published by the American Institute of Aeronautics and Astronautics, Inc. A computational study is performed for exploring flow and flame dynamics of a high-pressure hydrogen/oxygen coflow jet with the effects of post thickness and moment flux ratio. A two-dimensional model with a splitter plate, which represents a post configuration of an injector of rocket engines, is adapted to fully resolve the combustion flow field. The compressible Navier-Stokes equations are solved with a detailed chemical kinetic mechanism in a manner of direct numerical simulation. The result shows that the post thickness largely affects the temperature distribution in a recirculation region established behind the post. The temperature distribution is determined with the amount of incoming high-temperature combustion gas and unburned hydrogen gas, which significantly changes with the post thickness. The effect of the momentum flux ratio clearly appears in the case of thicker post configuration, while in the case of thinner post configuration no major differences are identified for all the momentum flux ratio. The study shows a tendency that thicker post geometries with smaller J numbers provide lower temperature fields in the recirculation region behind the post, thus preliminarily indicating some difficulty of maintaining a flame anchoring in the recirculation region.
  • 藤田晴彦, 伊藤祐太朗, 山田眞平, 下栗大右, 佐藤伴音, 寺島洋史, 河野通治, 本田雄哉, 植木義治, 横畑英明
    日本伝熱シンポジウム講演論文集(CD-ROM) 2017年
  • 榎尚也, 松永学, 高橋裕介, 寺島洋史, 大島伸行
    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講演集(CD-ROM) 2017年
  • 大橋達志, 松永学, 高橋裕介, 寺島洋史, 大島伸行
    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講演集(CD-ROM) 2017年
  • Takahide Araki, Daiki Muto, Hiroshi Terashima, Nobuyuki Tsuboi
    AIAA SciTech Forum - 55th AIAA Aerospace Sciences Meeting 2017年01月 
    © 2017 by Rolls-Royce North America Holdings Inc. Three-dimensional numerical simulations of coaxial nitrogen jet mixing under supercritical pressures are performed with an emphasis on the effects of pressure on the mixing. The results show that the overall mixing behaviors are similar in terms of normalized density profiles between 5 and 12 MPa. However, the normalized temperature profiles are different between two conditions because non-linear change of the density to the temperature near the critical point. The peak of the constant pressure specific heat is particularly observed in the case of 5 MPa. The density and temperature fluctuations become large between the inner jet flow and the outer jet flow. Although the density fluctuation distributions are similar between two cases, the larger fluctuation region of temperature is observed in the case of 5 MPa. This is because the temperature increases are different near the critical temperature between two pressure conditions.
  • 寺島洋史, 松木亮
    燃焼シンポジウム講演論文集 2016年11月
  • 佐藤伴音, 寺島洋史, 大島伸行
    燃焼シンポジウム講演論文集 2016年11月
  • 寺島洋史, 大門優
    日本機械学会熱工学コンファレンス講演論文集(CD-ROM) 2016年
  • 武藤大貴, 武藤大貴, 寺島洋史, 坪井伸幸
    数値流体力学シンポジウム講演論文集(CD-ROM) 2016年
  • Hiroumi Tani, Yutaka Umemura, Yu Daimon, Hiroshi Terashima, Mitsuo Koshi
    54th AIAA Aerospace Sciences Meeting 2016年01月 
    © 2016, American Institute of Aeronautics and Astronautics Inc, AIAA, All rights reserved. The vaporization and burning of the N 2 H 4 and NTO droplets were simulated with the interface-tracking method to accurately explore the auto-ignition processes and the flame structures. The N 2 H 4 vapor plume developed behind the N 2 H 4 droplet and reacted with the ambient NO 2 gas through the hydrogen abstraction reactions. Thus, the N 2 H 4 vapor and NO 2 gas mixtures behind the droplet were preheated and reached the auto-ignition at a few ms. The auto-ignition occurred in the multiple points almost at the same time. After the ignition, the premixed flame developed around the droplet. Thus, the vaporization of the liquid N 2 H 4 near the surface became significant. Then, the double flame structures which comprise the inner decomposition flame and oxidation flame appeared around the droplet. The NTO droplet was not auto-ignited in the computational time of the present study because little N 2 O4 vapor near a saturated temperature decomposed to NO 2 gas which is necessary for the hydrogen abstraction reactions. When the ignition was forced, the double flames developed. The outer decomposition flame propagated to the boundaries of the computational domain, while the inner oxidation flame appeared near the droplet. Except for the propagation of the decomposition flame, the NTO droplet combustion was similar to that of the industrial fuels.
  • Daiki Muto, Hiroshi Terashima, Nobuyuki Tsuboi
    54th AIAA Aerospace Sciences Meeting 2016年01月 
    © 2016, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. Three-dimensional numerical simulations of cryogenic coaxial jets under supercritical pressure are performed with flushed and recessed injectors to investigate the effect of a recess on the coaxial mixing. A hybrid ILES/RANS method is applied to simulate wallbounded injectors and a recessed region. The recessed injector enhances the density decay and the temperature increase on the central axis, indicating the improvement of mixing compared with the flushed injector. However, the mixing improvement by the recess is not significant in the present conditions. The recess also induces distinct vortex rings around the outer jet. The power spectra of the velocity fluctuations also demonstrated that the low-frequency velocity fluctuations are clearly induced by the recess which frequency corresponds to the large vortex structures.
  • Yu Daimon, Hiroumi Tani, Hiroshi Terashima, Mitsuo Koshi
    54th AIAA Aerospace Sciences Meeting 2016年01月 
    © 2016 by the American Institute of Aeronautics, and Astronautics, Inc. All rights reserved. Hydrazine (N 2 H 4 )/nitrogen dioxide (NO 2 ) un-like doublet impinging gas jets were simulated to explore the hypergolic ignition processes in a N 2 H 4 /N 2 O4 bipropellant thruster. The three-dimensional compressible Navier-Stokes equations with a detailed chemical kinetics mechanism, in which more than 200 chemical reactions were directly taken into account, were solved to reveal the influence of the chemical reaction. The differences of three-dimensional structures of hypergolic ignition process and mechanism of flame holding between the two different inlet gas temperatures of 400 and 600 K were discussed in order to investigate the influence of the induction time of chemical reaction on the three-dimensional flowfield. The computed results clarified that the ignition time of impinging gas jets can be significantly influenced by the ignition delay of the detailed chemical kinetics mechanism. In addition, the intermittent multi-ignitions played a significant role in the mechanism of flame holding.
  • H. Terashima, Y. Daimon
    52nd AIAA/SAE/ASEE Joint Propulsion Conference, 2016 2016年01月 
    © 2016, American Institute of Aeronautics and Astronautics Inc, AIAA. All rights reserved. A two-dimensional detailed numerical simulation is performed for combustion flow field of a GCH4/GOX single injector using detailed chemical kinetics with the compressible Navier-Stokes equations. A detailed mechanism of CH4, 33 chemical species and 150 re- actions, is efficiently and directly introduced. The result shows that the relatively high- temperature and CH4-rich recirculation region is established in the upper and lower corners of the combustion chamber. The result, with a at inlet profile, interestingly shows the generation of an unstable combustion mode, which is not observed with a smooth inlet pro- file. It is shown that the disappearance of non-premixed flames behind the GOX post is a trigger for the unstable combustion mode through the production of partly premixed gases and the generation of autoignition at several locations in the combustion chamber, which may be caused by the extent of the incursion of GCH4 and GOX jets in the recirculation region behind the GOX post.
  • Daiki Muto, Nobuyuki Tsuboi, Hiroshi Terashima
    53rd AIAA Aerospace Sciences Meeting 2015年01月 
    © 2015 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. The effects of the injector geometries on co-flowing planar cryogenic jet mixings under a supercritical condition are numerically investigated. The present study focuses the recess of the coaxial injector which is widely applied in practical liquid rocket engines. The present numerical method applies an ILES/RANS hybrid method to simulate the jet mixing in the wall-bounded recessed region. As a validation of the present method, a mono-planar jet and a round jet simulations are carried out, and the results agree well with an experimental result. To examine the effects of the recess length on the coaxial injections, two-dimensional co-planar jet simulations at the supercritical pressure are performed in three recess lengths. The recessed cases show the strong flapping motions of the high densty jet, and as a result, the injected fluids are mixed well compared with the case without the recess. While there is a small difference on the potential core length between the cases without the recess and the shorter recess, the longer recess case largely shortens the jet core.
  • Hiroumi Tani, Hiroshi Terashima, Ryoichi Kurose, Tomoaki Kitano, Mitsuo Koshi, Yu Daimon
    53rd AIAA Aerospace Sciences Meeting 2015年01月 
    © 2015 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved. Hydrazine (N 2 H 4 ) spray/gaseous nitrogen tetroxide (NTO) co-flowing jets were simulated to explore the hypergolic combustion flows in N 2 H 4 /NTO bipropellant thrusters. The Navier-Stokes equations with the use of a detailed chemical kinetics mechanism and dispersed droplets with evaporation models were solved in a manner of direct numerical simulations. The influence of the evaporation of N 2 H 4 droplets upon the hypergolic ignition processes and flame structures was investigated. Before the auto-ignition, the N 2 H 4 vapor and the ambient NTO gas mixtures were preheated as a result of the hydrogen abstraction reactions, whereas the evaporation of the N 2 H 4 droplets decreased the temperature of the gas mixtures. When the temperature of the NTO flows was sufficiently high, the N 2 H 4 vapor and NTO gas mixtures were preheated and could auto-ignite near the leading edge of the N 2 H 4 spray. This occurred because the heat transfer from the ambient gases to the droplets and the N 2 H 4 vapor enhanced the hydrogen abstraction reactions in the leading edge. After the auto-ignition, the double flame structure appeared, comprising the outer diffusion flames and inner decomposition flame. Interestingly, the inner decomposition flame and the N 2 H 4 vapor flow exhibited a sinusoidal behavior. This behavior was initiated by the locally expanded decomposition gases and developed by the supply of the N 2 H 4 droplets to the decomposition gases at relatively high temperatures. When the droplet size was small, the auto-ignition was not always enhanced because the temperature of the N 2 H 4 vapor and the NTO gas mixtures decreased. Furthermore, the sinusoidal behavior of the inner decomposition flame was less significant because the flame’s development depended on the temperature of the N 2 H 4 vapor and inertial mass of the droplets.
  • Hiroshi Terashima, Mitsuo Koshi
    52nd Aerospace Sciences Meeting 2014年01月 
    © 2014, American Institute of Aeronautics and Astronautics Inc. All rights reserved. Three-dimensional simulations for nitrogen/hydrogen mixing of a shear coaxial injector under supercritical pressures are conducted in order to investigate their mixing characteristics. A high-order numerical method based on a sixth-order compact scheme is applied. The present study parametrically covers two pressures of 4 and 8 MPa and two outer hydrogen jet conditions, while an inner nitrogen jet condition and a mixture ratio are kept to be same. The discussions on mean and fluctuation properties in addition to instantaneous flow fields are made in order to characterize mixing features for the coaxial injection. A colder outer hydrogen jet provides less mixing behavior compared to a warmer hydrogen jet, generating a longer dense-core of the inner nitrogen jet; unmixed nitrogen and hydrogen remain in downstream regions. Some unique features are found in the temperature profiles on the centerline of inner nitrogen jet as a flattened or a wave-like profile under 4 MPa, while no such unique features are not observed under 8 MPa. In the case of a warmer outer hydrogen jet, the hydrogen density show a maximum peak just after the injection because of the heat transfer to the inner nitrogen jet. The density and temperature fluctuation distributions are determined by the unique thermodynamic fluid variations around the critical temperature, and the velocity and temperature differences among two jets and chamber fluid. Finally, a clear dependence of dense-core length on the momentum flux ratio is demonstrated on comparison with experimental and computational data.
  • Akira Kikusato, Kazuya Kogo, Beini Zhou, Kusaka Jin, Yasuhiro Daisho, Kiyotaka Sato, Hidefumi Fujimoto, Hiroshi Terashima, Youhi Morii
    SAE Technical Papers 2014年01月 
    Copyright © 2014 SAE International. The objective of the present study is to analyze soot formation in diesel engine combustion by using multi-dimensional combustion simulations with a parallelized explicit ODE solver. Parallelized CHEMEQ2 was used to perform detailed chemical kinetics in KIVA-4 code. CHEMEQ2 is an explicit stiff ODE solver developed by Mott et al. which is known to be faster than traditional implicit ODE solvers, e.g., DVODE. In the present study, about eight times faster computation was achieved with CHEMEQ2 compared to DVODE when using a single thread. Further, by parallelizing CHEMEQ2 using OpenMP, the simulations could be run not only on calculation servers but also on desktop machines. The computation time decreases with the number of threads used. The parallelized CHEMEQ2 enabled combustion and emission characteristics, including detailed soot formation processes, to be predicted using KIVA-4 code with detailed chemical kinetics without the need for reducing the reaction mechanism. After validating the code, diesel engine combustion was simulated to investigate combustion and emission characteristics, focusing on soot formation, growth and oxidation at different EGR ratios. To predict soot formation, a gas-phase polycyclic aromatic hydrocarbons (PAH) precursor formation model was coupled with a detailed phenomenological particle formation model, which included soot nucleation from precursors, surface growth/oxidation and particle coagulation. The results indicate that increased soot emission at high EGR ratios is mainly caused by decreased oxidation by oxygen and OH radicals because mixing fuel and gases (including oxygen and OH) has significant effects on reducing the mass of soot.
  • Youhi Morii, Hiroshi Terashima, Mitsuo Koshi, Taro Shimizu, Eiji Shima
    50th AIAA/ASME/SAE/ASEE Joint Propulsion Conference 2014 2014年01月 
    © 2014 by authors. Published by the American Institute of Aeronautics and Astronautics, Inc. A simple yet robust and fast time integration method is proposed for efficiently solving stiff chemical kinetic ordinary differential equations. The proposed method is based on a general formula which preserves the conservation laws for any integration operators con- structed using the Lagrange multiplier method. A quasi-steady-state approxixmation is used as the integrator. The time step size is automatically controlled by using a Lagrange multiplier so that the error, which is caused by the Lagrange multiplier method, is small. The results of several ignition problems demonstrate the robustness and accuracy of the proposed method in comparison with other integration methods such as a implicit inte- gration method (VODE), a multi time-scalse method (MTS), and a modified CHEMEQ2. The proposed method, named ERENA, provides the fastest performance for the most of conditions used in this study.
  • Daiki Muto, Nobuyuki Tsuboi, Hiroshi Terashima
    52nd Aerospace Sciences Meeting 2014年01月 
    © 2014, American Institute of Aeronautics and Astronautics Inc. All rights reserved. Effects of recess on the mixing of the cryogenic coaxial jet under supercritical conditions were numerically investigated. Recessed injectors have been applied as coaxial injectors of liquid rocket engines because it is expected to improve propellant mixing and combustion efficiency. In the present study, a numerical simulation code was developed to examine jet mixing under the supercritical pressures. For validation of the present method, a two-dimensional planner jet simulation was carried out. The present result showed good agreement with an experimental result. As a preliminary study of coaxial jets, co-planner simulations were performed in a recessed case and an unrecessed case. Large-scale vortical structures were observed in the both cases, and the recessed case showed the larger struc- tures relative to the unrecessed case. This is because of the destabilization mechanism of a recessed region. These instabilities improved entrainments of ambient uid, and as a result, the enhanced jet mixing shortens the jet-core length.
  • Yu Daimon, Hiroshi Terashima, Mitsuo Koshi
    51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition 2013 2013年08月 
    A detailed chemical kinetic mechanism for hypergolic ignition of N2H4/NO2 gas mixture at low temperatures has been constructed. In this mechanism, the hypergolic ignition is caused by following sequential reactions of H atom abstraction from N2Hm by NO2. N2H4+ NO2= N2H3+ HONO or HNO2(R1) N2H3+ NO2= N2H2+ HONO (R2) N2H2+ NO2= N2H + HONO (R3) N2H + NO2= N2+ HONO (R4) These reactions are exothermic, especially heat release by the reaction (R4) is large because of N2 production. Temperature rise caused by the heat release accelerates the initiation reaction (R1). This 'thermal feedback' is responsible to the hypergolic ignition at ambient temperatures. Since no experimental and theoretical information was available on these reactions, rate coefficients were evaluated on the basis of transition state theory, uni-molecular rate theory, and master equation analysis with quantum chemical calculations of potential energy curves. Results of simulations by using the present mechanism including reactions (R1)-(R4) reasonably agree with existing experimental data. © 2013 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.
  • Hiroshi Terashima, Mitsuo Koshi
    51st AIAA Aerospace Sciences Meeting including the New Horizons Forum and Aerospace Exposition 2013 2013年08月 
    Two-dimensional planar and three-dimensional round nitrogen jets under supercritical pressures are simulated with a wide range of conditions in order to explore its unique characteristics. A high-order method using a sixth-order compact scheme is applied. The present study parametrically covers two supercritical pressures of 4 MPa and 8 MPa and three jet injection temperatures between a cryogenic jet of 82 K and a warmer jet of 133 K. Unique characteristics are found in both the mean temperature and the temperature fluctuation profiles on the centerline, i.e., slower increase of jet temperature and relatively weak temperature fluctuation, only in case of a near-critical pressure of 4 MPa and a cryogenic jet of 82 K. The temperature profile on the centerline of the cryogenic jet of 82 K under 4 MPa uniquely consists of four characteristic regions. The other conditions show no distinct features. The present study suggests that the distributions of the specific heat capacity at constant pressure help to explain the generation of the unique characteristics on the temperature. © 2013 by the authors.
  • Hiroshi Terashima, Soshi Kawai, Mitsuo Koshi
    American Society of Mechanical Engineers, Fluids Engineering Division (Publication) FEDSM 2012年12月 
    We present an interface-capturing method for fluid interfaces in compressible multicomponent flows using high-order central-difference-based schemes. Numerical diffusion terms are consistently designed so that the velocity, pressure, and temperature equilibriums are maintained at the fluid interfaces, while serving as an efficient interface-capturing. Advection problems of a contact discontinuity and a material interface shows that 1) the present method maintains the velocity, pressure, and temperature equilibriums at the fluid interfaces (oscillation-free property) and 2) the numerical diffusion terms effectively works for suppressing spurious wiggles of the density or temperature. Comparisons with a conventional fully-conservative approach demonstrates the superiority of the present method in avoiding spurious oscillations. A shock tube problem of two-component gases shows the capability for capturing the shock wave while the velocity and pressure equilibriums are successfully maintained at the contact discontinuity. Copyright © 2012 by ASME.
  • Hiroshi Terashima, Soshi Kawai, Nobuhiro Yamanishi
    41st AIAA Fluid Dynamics Conference and Exhibit 2011年12月 
    Two-dimensional planar nitrogen jets in supercritical thermodynamic conditions are simulated using a high-resolution numerical method (which consists of a sixth-order compact difference scheme and a localized artificial diffusivity method) with the aim at exploring its unique characteristics. Effects of injection temperature, chamber pressure, and the equation of state on supercritical jet behaviors are investigated. Throughout the investigations, two major unique characteristics are found under the transcritical conditions. One unique characteristic are found in the mean temperature profile, as the slower increase of jet temperature, in the case of transcritical injections. The cause for the unique feature is simply explained by the corresponding temperature-density (T-ρ) diagrams. Another unique feature of supercritical jet flows appears in the generation of different flow scales. The power spectrums and the flow fields quantitatively and qualitatively show the different features of flow scales due to the injection conditions. In the transcritical injection, the production of smaller flow scales is considerably enhanced relative to the other injection cases, due to not only its higher density ratio, but also its abrupt variations (leap) of δρ/δT in the T-ρ diagrams. This study indicates, for the present conditions used, that the unique characteristics of supercritical jet flows appear in the mean temperature distributions and the generation of different flow scales, which are simply yet effectively explained by the T-ρ diagrams. © 2011 by the authors.
  • 寺島 洋史, TRYGGVASON Gretar
    微粒化シンポジウム講演論文集 = Symposium (ILASS-Japan) on Atomization 2009年12月
  • Hiroshi Terashima, Kenji Ono
    Computational Fluid Dynamics 2006 - Proceedings of the Fourth International Conference on Computational Fluid Dynamics, ICCFD 2006 2009年01月
  • Hiroshi Terashima, Kenji Ono
    2007 Proceedings of the 5th Joint ASME/JSME Fluids Engineering Summer Conference, FEDSM 2007 2007年12月 
    A compressible flow solver coupled with moving/deformed geometries on Cartesian grid with Signed Distance Field (SDF) is developed and its capability is investigated through computations of several basic flow fields for future applications with certain reliability. The flow solver is designed so that SDF includes sufficient geometrical information to compute flow fields. Since information of moving/deformed geometries is recognized as a change of the SDF between time steps, the flow solver can be coupled with moving/deformed geometries naturally. The implementation of this solver is simple and easy. No modification is needed in the main part of the flow solver. Furthermore, the interpolation and the corresponding stencils searching process are not required. Several basic flow fields around fixed/moving cylinders and a fixed sphere are computed in order to validate the proposed solver, in which the computed results are compared with available numerical and experimental results. The results demonstrated the method's capability for moderate Reynolds number flows around both of fixed and moving geometries. Based on the results, some criteria and problems for obtaining reliable solution are suggested. Copyright © 2007 by ASME.
  • Hiroshi Terashima, Kozo Fujii
    34th AIAA Fluid Dynamics Conference and Exhibit 2004年12月 
    Transonic and supersonic flutter characteristics of a delta wing configuration with external stores are simulated using the fluid/structure coupling method and the effect of the store aerodynamics is investigated. A delta wing with one and two external stores are considered in this study. The stores are located in the rearward of the wing simulating typical high speed aircraft configurations. The computational result shows that the flutter dynamic pressures over all the Mach numbers are reduced by adding external stores for both cases. In addition, the flutter dynamic pressures become much lower when the number of attached external stores is increased. In the case of the single external store, influence of the store aerodynamics can be divided into two regions according to the Mach number. It is found that neglecting the store aerodynamics leads to the overestimation of the flutter dynamic pressures in the supersonic region. In case of two external stores, the store aerodynamics influence on the flutter boundary only at M∞ = 1.22 unlike the case of the single external store. It is shown that the shock wave ahead of the external store influences on the pressure distributions on the lower surface and the location of the shock wave is important for the decision of the flutter boundary. © 2004 by the American Institute of Aeronautics and Astronautics, Inc.
  • 寺島 洋史, 藤井 孝藏
    年次大会講演論文集 : JSME annual meeting 2002年09月 
    Flutter phenomena for the AGARD445.6 standard aeroelastic wing in the transonic flow are simulated using the fluid/structure coupling method. The effect of time accuracy, number of the inner-iterations and comparison of a fully implicit coupling method with a loosely coupling method are examined. The sufficient number of inner-iterations in time is required for the accurate prediction of time responses when using a large time step. Improvement of the solution time accuracy for the aerodynamic equations is key for accurate aeroelastic computations and the one-step time lag generated in a loosely coupling method does not have an effect in the flutter prediction in transonic flows.
  • 藤井 孝蔵, 守屋 公一郎, 寺島 洋史, 堀江 俊行
    流体工学部門講演会講演論文集 2001年09月 
    CFD simulations using three-dimensional compressible Navier-Stokes equations are carried out for the establishment of the reliability of the CFD tools for the aerodynamic evaluation of space transportation systems. Simulations for Apollo-type capsule show that the simulations with typical number of grid points can give us reasonable result and CFD analysis can be a useful tool for the initial estimation of aerodynamic characteristics. Simulations for the FTB configuration show the dependency of the aerodynamic data on the order of accuracy of the simulations. The flow field near the base alters the characteristics. The example for the delta and double-delta wings which may be a candidate for the future TSTO systems shows that the success of the CFD simulations for delta wings does not necessarily justify the CFD results for similar configurations.
  • Satoru Yamamoto, Hiroshi Terashima
    Fluids 2000 Conference and Exhibit 2000年12月 
    Unsteady three-dimensional hypersonic shock/shock interference flows measured by Berry and Nowak1 are calculated using the shock-vortex capturing method with and without thermochemical nonequilibrium effect. This method contains the 4th-order compact MUSCL TVD scheme, the maximum 2nd-order LU-SGS scheme, and the AUSM-based scheme. Two typical 3-D flows around the forward swept fin at the angle of 0° and -15° are first calculated without reaction. The flow pattern and the shock heating on the body surface are well compared with the experiments. These 3-D flows specified by a high temperature condition are also calculated for resolving the effect of thermochemical nonequilibrium. The two-temperature model based on the Park model is employed for the chemical reaction model. The calculated results show that the increase of temperature changes the type of shock interference because of decreasing the shock standoff distance and also suggest that the shock heating might be very sensitive to the type of shock interference. Finally, unsteady flow characteristics captured both with and without thermochemical nonequilibrium effect are preliminarily expressed. © 2000 by the American Institute of Aeronautics and Astronautics, Inc. All rights reserved.

所属学協会

  • 米国航空宇宙学会   日本流体力学会   日本燃焼学会   日本機械学会   日本航空宇宙学会   

共同研究・競争的資金等の研究課題

  • 日本学術振興会:科学研究費助成事業 基盤研究(B)
    研究期間 : 2021年04月 -2024年03月 
    代表者 : 寺島 洋史, 河合 宗司
     
    超臨界圧燃焼流れシミュレーション(超臨界燃焼CFD)を実施するため,まず,熱,輸送物性,そして化学反応モデルに非理想性を考慮した流体物性モデルの開発を行った.非理想性を考慮した物性算出プログラムは,世界で標準的に使用される熱・輸送物性ライブラリChemkinと互換性を持つように設計し,任意反応機構を用いた超臨界燃焼CFD解析が可能となっている.特に本研究では,多くの関連研究で無視されてきた化学反応における非理想性を考慮したモデル構築を行った.化学反応における非理想性は,反応前後のギブス自由エネルギー変化をフガシティーで記述し,最終的には平衡定数の算出において考慮される.高圧水素火炎伝播実験と比較を行い,本モデルが実験で観測された質量燃焼速度の圧力負依存性を再現できることを示した.高圧条件においては,質量燃焼速度予測に理想気体モデルとの差が発生するが,この原因をJoule-Thomson(J-T)効果(エンタルピーの圧力依存性)で説明できることを提示した.例えばアルゴンで希釈されていれば,エンタルピーが負の圧力依存性を持つため,非理想モデルは,理想モデルに比べ低い層流燃焼速度および質量燃焼速度を予測する.J-T効果を用いた非理想性効果の議論はこれまでほとんどなく,本成果は,採択が約35%と競争的で知られている国際燃焼シンポジウムの口頭発表に採択されている.化学反応項LESモデル開発については,これまで開発してきた火炎モデル(LTF)をベースとし,火炎伸張を考慮したモデル拡張(LTF-Beta)を実施した.乱流平面火炎干渉場の解析を通して,wrinkled flame領域では,LTF-BetaがLTFに比べ直接数値解析に近い結果を予測できることを示した.異なる燃焼領域への適用や燃焼速度と火炎伸張率の線形モデルに対するさらなる理論考察が必要であり,継続して研究を実施する.
  • 日本学術振興会:科学研究費助成事業 国際共同研究加速基金(国際共同研究強化(A))
    研究期間 : 2022年 -2024年 
    代表者 : 寺島 洋史
  • 日本学術振興会:科学研究費助成事業 挑戦的研究(萌芽)
    研究期間 : 2019年06月 -2022年03月 
    代表者 : 河合 宗司, 寺島 洋史
     
    本研究では、火炎帯厚さを格子で解像しない粗い格子を用いても乱流燃焼現象を高忠実に再現可能な詳細反応機構に基づく乱流燃焼LES手法の確立を目指し研究を進めた。本研究ではまず火炎帯内部の化学種分布は変えずに空間方向に引き伸ばすことで(火炎帯を人工的に引き伸ばすことで)、粗い格子でも層流伝播火炎の燃焼速度や自着火を正しく再現する火炎帯拡大モデル、Localized thickened flame(LTF)モデルを提案した。また本提案LTFモデルを基盤とした詳細反応機構に基づく乱流燃焼LESモデルについて、火炎面曲率効果のモデリングを提案し、その有効性を検証した。
  • 日本学術振興会:科学研究費助成事業 基盤研究(C)
    研究期間 : 2017年04月 -2020年03月 
    代表者 : 寺島 洋史
     
    本研究では,ロケットエンジンなどに見られる臨界圧を超えた超臨界状態を対象として,多成分流体および燃焼現象の理解に資する数値シミュレーション技術を新たに開発した.超臨界流体では流体熱物性および輸送物性の非理想性効果が表れるが,それぞれ適切な非理想モデルを導入し,十分な検証を行うことで技術を確立させた.超臨界圧燃焼現象においては,特に拡散係数の影響が大きく,非理想モデル選択に注意が必要であることを明確にした.また,超臨界圧環境では,伝播火炎帯が非常に薄くなるため,高い計算格子解像度が必要となる.そこで,粗い計算格子でも火炎特性(火炎伝播および自着火現象)を再現できる新たな火炎モデル構築に成功した.
  • 日本学術振興会:科学研究費助成事業 基盤研究(B)
    研究期間 : 2015年04月 -2018年03月 
    代表者 : 桑名 一徳, 寺島 洋史, 林 直樹
     
    火災・爆発シミュレーションに特化した燃焼反応モデルを整備し、火災・爆発シミュレーションの精度向上を図ることを目的とし研究を実施した。まず、実験および1次元詳細反応シミュレーションの結果をもとに総括反応モデルの速度パラメータを導出した。得られた総括反応モデルやほかの簡易反応モデルを用いた数値シミュレーションを実施し、実験結果や詳細反応シミュレーションの結果と比較した。これにより、反応モデルの妥当性を検証した。また、拡散火炎と予混合火炎の性質を併せ持つトリプルフレームを対象とし、状況に応じて最適な反応モデルを選択できる方法論について検討を行った。
  • 日本学術振興会:科学研究費助成事業 基盤研究(B)
    研究期間 : 2015年04月 -2018年03月 
    代表者 : 坪井 伸幸, 寺本 進, 越 光男, 林 光一, 徳増 崇, 津田 伸一, 寺島 洋史, 清水 太郎, 朝原 誠, 森井 雄飛, 永島 浩樹, 武藤 大貴, 小澤 晃平
     
    高圧条件下での超臨界燃焼流体に対する熱力学的特性・流体力学的特性を明らかにするために,(1)単成分流体解析(2)多成分流体解析(3)実験 を実施した.単成分流体解析では,ILES/RANS hybrid法を用いて同軸の極低温超臨界窒素噴流を解析し,平均内側噴流長さについて過去の実験や数値計算との比較により妥当な結果を得ることができた.多成分流体解析については,エネルギー方程式と圧力発展方程式の両方を解くハイブリッド法を開発し,圧力振動を抑制することが可能となった.実験については,冷却チャネル内を流れる超臨界・亜臨界条件下の極低温窒素流について可視化画像や温度分布などを取得した.
  • 日本学術振興会:科学研究費助成事業 基盤研究(B)
    研究期間 : 2014年04月 -2018年03月 
    代表者 : 金子 成彦, 寺島 洋史, 山崎 由大
     
    バイオ燃料の活用やエコモビリティの実現に向けては,小型エネルギー変換機器の高効率化と環境性能の両立は必須条件である。本研究では,①天然ガス,バイオガス,石炭ガス等を燃料とする小型発電用ガスタービン,②バイオガスと液体燃料の両方を燃料とするデュアルフューエル自動車用レシプロエンジン,③ジェット燃料やバイオ燃料を燃料とする小型ターボジェットエンジンを対象に,高効率化と環境性能の両立実現の障壁となっている①燃焼振動対策,②計算負荷の軽いリアルタイム制御アルゴリズム,③ノズルから放出される騒音低減デバイス開発について,化学反応を絡めた新しい視点から各機関に内在する動的問題の解決策を提案した。
  • 日本学術振興会:科学研究費助成事業 挑戦的萌芽研究
    研究期間 : 2015年04月 -2017年03月 
    代表者 : 河合 宗司, 寺島 洋史
     
    詳細反応機構に基づく実燃焼流問題のシミュレーションを可能とする高精度かつ堅牢な数値計算法の確立を目的とし研究を行った。人工的に火炎面を厚くすることで粗い計算格子でも火炎を解像できるようにするArtificially Thickened Flame(ATF)モデルのアイディアを基にし、独自の視点に基づく物理的に矛盾無く支配方程式レベルで工夫する理論を詳細反応機構に基づく燃焼流れに発展させた。また今後の本手法の高次精度計算手法への展開を考慮し、化学種の保存則を丸め誤差レベルで常に満足する高次精度陰的フィルタースキームの開発を行った。
  • 文部科学省:科学研究費補助金(基盤研究(C))
    研究期間 : 2014年 -2016年 
    代表者 : 寺島 洋史, 越 光男, 森井 雄飛
     
    大規模詳細反応機構を用いた反応性流体シミュレーションを効率的に行う解析手法を提案した.提案手法は,硬直性の強い化学反応方程式に対する高速陽的時間積分法と多成分流体輸送係数に対する化学種バンドル法から構成されている.それぞれの手法は,従来法に対し,解の精度を大きく損ねることなく,2桁から3桁の高速化を可能とした.提案手法は,自動車ノッキング現象,高圧ロケットエンジン燃焼現象,予混合火炎不安定性現象と,幅広い燃焼現象に適用され,新しい知見の創出に貢献した.
  • 文部科学省:科学研究費補助金(基盤研究(A))
    研究期間 : 2011年 -2014年 
    代表者 : 土橋 律, 越 光男, 桑名 一徳, 茂木 俊夫
     
    ①爆発の発生限界(爆発限界)の予測手法の確立昨年度までに開発したN2H4/NTO混合気体の自己着火性を記述できる詳細化学反応機構に基いて,層流燃焼速度および着火遅れ時間の温度,圧力依存性に関するシミュレーションを実施した.得られた結果を用いて,N2H4/NTO混合気体の爆発限界予測を行った.②爆発被害の影響度評価手法の確立(a)火炎の不安定性解析: 爆発シミュレーションの精度を向上するために,不安定性による火炎の加速を再現できる総括反応モデルを構築した.具体的な手法としては,まず,火炎伸長が存在する場での一次元的な火炎構造を詳細反応モデルを用いた数値シミュレーションにより求め,燃焼速度の火炎伸長に対する応答性を調べた.そして,この応答性を再現できるように総括反応モデルの速度パラメータを決定した.本研究で提案する総括反応モデルを用いれば,火炎の不安定性による伝播速度の上昇を精度よく再現でき,ガス爆発事故の被害予測の精度向上につながると期待される.(b)爆発実験: 昨年度取得したデータの再現性の確認および実験データの精度向上を目的として,3m立法(27m3)のガス爆発実験を実施した.得られた火炎伝ぱに関するデータを下に流体学的不安定性の影響について調べた.火炎伝ぱの加速指数はPeclet 数が大きくなるに従って増加し,ある限界値に至る.つまり,火炎の不安定性による火炎伝ぱの加速現象は大スケールでは自己相似的に発達することが明らかになった.
  • 文部科学省:科学研究費補助金(基盤研究(B))
    研究期間 : 2011年 -2013年 
    代表者 : 越 光男
     
    自動車エンジンのノッキングやガスタービンの振動燃焼の解析には化学反応方程式と流体力学方程式を同時に解く事が必要であるが、燃焼反応には数百もの化学種が関与するためにこのようなシミュレーションは不可能であった。これを可能とするために、燃焼反応の特性を損なわずに解析に必要な変数の数を大幅に減少させる方法を開発した。また反応法的式の超高速積分法も開発した。これにより、化学反応と流体現象がカップリングする現象の解析が初めて可能となった。
  • 文部科学省:科学研究費補助金(基盤研究(B))
    研究期間 : 2011年 -2013年 
    代表者 : 坪井 伸幸, 寺本 進, 越 光男, 林 光一, 徳増 崇, 津田 伸一, 清水 和弥, 清水 太郎, 谷 洋海, 朝原 誠, 森井 雄飛
     
    高圧条件下での超臨界燃焼流体に対する熱力学的特性・流体力学的特性を明らかにするために,(1)RANS解析(2)LES解析(3)実験 を実施した.RANS解析では,単軸の極低温超臨界窒素噴流の密度分布について実験結果と比較し,妥当な結果が得られることを確認した.さらに,多化学種の質量保存式を含む流体対応の前処理法を開発し,非定常な低速酸水素剪断流における剪断層の不安定現象を捉えることに成功した.LES解析については,遷臨界混合層において擬臨界温度が乱流渦構造に与える影響は小さかった.実験については,極低温遷臨界窒素噴流における中心軸上の温度勾配は,擬臨界温度付近で緩やかになる傾向が見られた.
  • 文部科学省:科学研究費補助金(若手研究(B))
    研究期間 : 2011年 -2012年 
    代表者 : 寺島 洋史
     
    本研究では,高速流中の推進剤混合過程解明を目指し,衝撃波を含む圧縮性高速流中の流体界面挙動に対するマルチスケールシミュレーション手法の開発とその流体不安定メカニズム解明に関する研究を行った.高速流中で生じる衝撃波,それと干渉する流体界面挙動をシミュレーションするため,圧縮性流体方程式を用い,界面追跡にフロントトラッキング法を,界面境界条件にゴーストフルイド法を採用した解析法を構築した.これまでの手法に対して,界面において厳密リーマン解を求めるよう大幅な修正を施し,界面マーカー点における物理量を背景格子点に埋め込む一般的な変換手法を構築した.また,液滴崩壊により追跡できなくなったマーカー点を粒子の運動方程式へつながるフレームを検討した.密度比1000倍となる気液界面問題に適用し,実験や過去の数値解析結果との比較から精度良い結果が得られることを示した.衝撃波と液滴の干渉問題では,従来捕えることが難しいとされた液滴のストリッピングを含めた液滴崩壊現象を比較的少ない格子点で捕えることに成功した.
  • 文部科学省:科学研究費補助金(若手研究(B))
    研究期間 : 2006年 -2008年 
    代表者 : 寺島 洋史
     
    本研究は,AirbagやParachuteといった流体と干渉する薄膜展開構造物のダイナミクスシミュレーションを行う数値手法の開発と設計への応用を目的とする.流体解析には,大変形薄膜物体を容易かつロバストに扱うため直交格子を用い,構造解析には幾何学的非線形性を考慮した有限要素法を採用した.今年度は,直交格子を用いた流体解析手法の開発と,流体と移動剛体が連成した流体剛体連成解析手法を開発し,その信頼性を評価した.本手法は以下の特長を持つ:1)距離関数場(Signed Distance Filed : SDF)を使用2)プログラムの修正は境界条件部分のみで,オリジナルソルバーからの修正が非常に容易3)鏡面境界条件で見られる境界付近の物理量内挿処理とそれに伴う点探索処理が無い4)SDF情報の変化がダイレクトに物体移動の情報を表し,流体構造連成問題を合理的に扱える特に3)は移動変形物体を扱う際に有利な点である.直交格子の解析でしばしば問題となるのが計算結果の信頼性である.信頼性を検証するため,基本的な流れ場(円柱,球,振動円柱)のシミュレーションを行い,実験や他の計算と比較を行った.固定円柱と球の計算結果から,レイノルズ数500までではあるが,他の結果と良い一致を得た.また,格子解像度の影響を調べ,予想される境界層厚さに10点程度の格子点を配置することにより,信頼性ある解が得られることを示した.これは今後の解析における格子解像度選択の指標となる.振動円柱の解析では,SDFの変化をベースとした本手法が移動物体のシミュレーションに対して有効であることを示し,また,固定円柱の流れ場に比べ,力の振幅や位相を正確に求める上で格子解像度が重要であることを明らかにした.また,粗い格子を用いると,力の時間履歴に虚偽振動が発生することがわかった.この原因は今後の課題である.


Copyright © MEDIA FUSION Co.,Ltd. All rights reserved.