فهرست مطالب

Applied Fluid Mechanics - Volume:11 Issue: 2, Mar-Apr 2018

Journal Of Applied Fluid Mechanics
Volume:11 Issue: 2, Mar-Apr 2018

  • تاریخ انتشار: 1396/12/05
  • تعداد عناوین: 21
|
  • J. K. Arthur * Pages 297-307
    One of the essential areas of the study of transport in porous medium is the flow phenomena at the onset of inertia. While this area has attracted considerable research interest, many fundamental questions remain. Such questions relate to things such as the nature of the multi-dimensional velocities of the flow, the evolution of inertia, the differences in flow phenomena at various complexity of porous media, and the best constitutive equation for the flow. To resolve some of these questions, the present research program was designed to experimentally investigate pressure-driven flow through two- and three-dimensional porous media at the onset of inertia. Specifically, the goals in view were to obtain velocity data and pressure measurements, apply the benchmark experimental data to study the evolution of inertia, distinguish differences in such evolution with respect to the parameters of the porous media, and to establish the constitutive equation that best describes the porous media flow when inertia sets in. What particularly sets this work apart, is the use of particle image velocimetry (PIV) – an experimental technique that captures multi-dimensional flow quantities, as opposed to mere flow rates. Using PIV then, detailed velocity measurements were conducted for flows through model porous media of solid volume fraction 6%, 12%, and 22%. The velocities were spatially averaged to obtain average streamwise and transverse components. In addition to the velocity measurements, differential pressure measurements were obtained using pressure-measurement gauges and transducers. The pressure and velocity data sets were then statistically analyzed and presented to provide a complete set of experimental data to characterize the flow through the model porous media. The results show that the velocity flow domain is dictated by the streamwise velocities, which are at least an order of magnitude greater than the transverse components. Furthermore, pressure drag was found to increase with compactness and complexity of the porous media. While inertia increases exponentially from particle Reynolds number ~ 1 – 3 onwards, it is apparently subdued by the form drag that tends to dominate the flow through complex media. Overall, the flow at the onset of inertia is best described by a power law. These results provide insights that are applicable to flows such as those near well bores and fractures where seepage velocities are relatively high.
    Keywords: Particle image velocimetry, Porous media, Darcy, Forchheimer, Izbash, Inertia
  • N. Reddy, K. Murugesan Pages 309-322
    Double diffusive convection (DDC) flows are widely seen in many industrial processes where the thermo-solutal buoyancy forces generates vorticity and initiates convective heat and mass transfer. In this paper numerical computations are conducted on this behaviour inside cavities of different aspect ratio at nominal Rayleigh number using a finite element based code. Velocity –vorticity form of Navier-Stokes equations are solved along with energy and solutal concentration conservation equations simultaneously using Galerkin’s weighted residual method. Bottom wall is assumed hot and salted while top wall is maintained as sink, both side walls of the cavity are assumed to be adiabatic to heat and mass flow. Generally cavities with the present boundary conditions exhibit weak vorticity and convection characteristic especially at low Rayleigh number. In this numerical work an attempt is made to explore the role of variation in relative strength of thermal and solutal buoyancy forces on flow characteristics and mode of heat and mass transfer in such conditions. Simulation results have been reported for different buoyancy ratios in the range -2≤N≤2 , Rayleigh number varying from 1.0e5 to 1.0e3, for cavities of aspect ratios, 0.5(shallow), 1 (square) and 2 (deep). Flow contours are well validated with the results in the literature. The fluid rotation patterns are captured and reported under different operating conditions chosen, the vorticity generation is observed relatively low for deep cavity when compared to other two. Investigations revealed that fluid convection gets greatly hampered when operated in negative buoyancy ratio regime and require relatively higher Rayleigh number to change the mode of heat transfer from diffusion to convection.
    Keywords: Double diffusive convection, Thermal, solutal buoyancy forces, Vorticity, Bottom heated cavity
  • E. Fadaei Kermani, G. A. Barani, M. Ghaeini-Hessaroeyeh Pages 323-329
    The present paper deals with a numerical method for prediction of cavitation damage level and location on dam spillways. A method was applied to predict the intensity of cavitation damage to spillways, using the fuzzy k-nearest neighbor algorithm. Five levels of damage intensity were considered to predict cavitation damage in the spillway of Karun-1 Dam in Iran. According to the results, the proposed model could properly predict the location and intensity of damage in comparison with the actual damage reports of past floods. According to the Pearson's correlation coefficient, mean absolute error, coefficient of residual mass, and normalized root mean square error, the fuzzy k-nearest neighbor model is efficient and suitable.
    Keywords: Cavitation damage, Spillways, Fuzzy k-nearest neighbor model, Damage intensity
  • S. H. I. Jingwei, W. A. N. G. Zhanxue, Z. H. O. U. Li, S. U. N. Xiaolin Pages 331-342
    Shock vectoring control (SVC) is an important method of fluidic thrust vectoring (FTV) for aero-engine exhaust system. It behaves better on nozzle of high pressure ratio, and is considered as an alternative TV technology for a future aero-engine with high thrust-to-weight ratio. In this paper, the flow mechanism and vector performance, including the vector angle (δp) and thrust coefficient (Cfg), of 2D and axisymmetric SVC nozzles were investigated after the validation of turbulence models by experimental data. The influence of aerodynamic parameters, e.g. nozzle pressure ratio (NPR), secondary pressure ratio (SPR) and free-stream Ma number (M∞) on flow characteristics and vector performance were studied numerically, and results show that unbalanced pressure distributions on nozzle internal walls determine δp, while shock waves dominate thrust loss, referring to Cfg. The “pressure release mechanism” of an axisymmetric SVC nozzle causes vector angle about 16.54% smaller than that of a 2D SVC nozzle at NPR of 6. The induced shock wave interacts with nozzle upper wall at SPR of 1.5, and results in the δp of a 2D SVC nozzle 12% smaller. A new parameter (Fy,modi) of side-force was redefined for free-stream conditions, taking the pressure distributions on nozzle external walls into account. Results indicate that pressure connection on nozzle external walls of an axisymmetric SVC nozzle causes vector performance better at M∞ >0.3 and the δp is about 11.2% larger at transonic conditions of M∞ of 0.9 and 1.1.
    Keywords: SVC, Transverse injection, Vector performance, Aerodynamic parameters, Flow control
  • K. Mostafapour, N. M. Nouri, M. Zeinali Pages 343-352
    The hydrodynamic characteristics of autonomous underwater vehicles (AUVs) play a significant role in the design and analysis of their maneuverability. This paper evaluates the effects of the Reynolds (Re) number on the hydrodynamic characteristics of AUV for various angles of attack (AOA). To estimate the hydrodynamic parameters, a numerical modelling based on computational fluid dynamics (CFD) is employed. Reynolds numbers between 2106 and 150106 were examined at -10º to 10º AOAs. Experimental tests for the same AUV in Re = 2106 in the water tunnel were carried out for CFD validation. A comparison of the results showed an acceptable agreement between the numerical method and the experimental results. The results show that hydrodynamic parameters can be a function of Re and converge on a constant in a limited value when the Re number increases. Results of independent parameters, can be used for full-scale without the establishment of dynamic similarity.
    Keywords: Hydrodynamic characteristics, AUV, CFD, Water tunnel
  • B. Kundu, K. S. Lee Pages 353-359
    In this study, a mathematical analysis by considering the effect of an actual index of expansion clearly shows a persistent of the existence of subsonic flow after the throat to a down stream in the region of divergent part to produce a supersonic velocity at the exit of a convergent-divergent nozzle. The length of the divergent part where subsonic velocity found is dependent upon the magnitude of the nozzle efficiency and the actual index of expansion. The change in velocity from subsonic, sonic and supersonic occurs only in the divergent part while the corresponding frictionless behavior has the classical features (subsonic in the convergent, sonic at the throat, and supersonic in the divergent). This is mainly due to thermodynamic processes which result a change in enthalpy due to friction and a gain in entropy. The reference conditions are newly derived for an actual frictional flow condition. This design aspect differs in a physical manner corresponding to that from an isentropic flow. An actual nozzle shape for convergent-divergent nozzles is also investigated under a non-isentropic flow condition.
    Keywords: Actual shape, Convergent, divergent nozzle, Friction, Reference condition, Subsonic velocity
  • F. Forghany, M. Taeibi-Rahni, A. Asadollahi Ghohieh Pages 361-374
    The present study attempted to utilize a computational investigation to optimize the external freestream flow influence on thrust-vector control. The external flow with different Mach numbers from 0.05 to 1.1 and with optimum injection angles from 60˚ to 120˚ were studied at variable flow conditions. Simulation of a converging-diverging nozzle with shock-vector control method was performed, using the unsteady Reynolds- averaged Navier-Stokes approach with Spalart-Allmaras turbulence model. This research established that freestream flow and fluidic-injection angle are the significant parameters on shock-vector control performance. Computational results indicate that, increasing freestream Mach number would decline the thrust vectoring effectiveness. Also, optimizing injection angle would reduce the negative effect of external freestream flow on thrust-vector control. Moreover, increasing secondary to primary total pressure ratios and decreasing nozzle pressure ratios at different freestream Mach number would decrease dynamic response of starting thrust-vector control. Additionally, to lead the improvement of the next generation of jet engine concepts, the current study aimed to originate a database of variable external flow with effective aerodynamic parameters, which have influence on fluidic thrust-vector control.
    Keywords: Shock vector control, Freestream flow, Fluidic, injection angle, Dynamic response
  • A. Olcay, A. Amindari, K. Kirkkopru, H. Yalcin Pages 375-384
    The aortic valve is located at left ventricular outlet and is exposed to the highest pressure in the cardiovascular system. Problems associated with the valve leaflet movement can cause complications for the heart. Specifically, aortic stenosis (AS) arises when aortic leaflets do not efficiently open. In the present study, Lagrangian Coherent Structures (LCSs) were utilized by processing a variety of Computational Fluid Dynamics (CFD) models velocity vector data further to identify the characteristics of AS jets. Particularly, effective orifice areas (EOA) for different cases were accurately identified from unstable manifolds of finite time Lyapunov exponent (FTLE) fields. Calcified leaflets were modeled by setting the leaflet's Young modulus to 10 MPa and 20 MPa for moderately and severely calcified leaflets respectively while a healthy leaflet's Young modulus was assigned to be 2 MPa. Increase in calcification degree of the leaflet caused destruction of the vortex structures near the fibrosa layer of the leaflet indicating a malfunctioning for the movement mechanism of the leaflet. Furthermore, when we analyzed stable manifolds, we identified a blockage region at the flow upstream due to the stagnant blood here. Compared to a healthy case, for the calcified valve, this blockage region was enlarged, implying an increase in AS jet velocity and wall shear stress on leaflets. As a conclusion, results from the present study indicate that aortic leaflet malfunctioning could be accurately evaluated when LCS technique was employed by post processing velocity vector data from CFD. Such precise analysis is not possible using the Eulerian CFD approach or a Doppler echocardiography since these methods are based on only analyzing instantaneous flow quantities and they overlook fluid flow characteristics of highly unsteady flows.
    Keywords: Lagrangian coherent structures, Aortic valve, Stenosis, Computational fluid dynamics, Fluid structure interaction, Calcification, Hemodynamics, Vortex, Wall shear stress, Pressure
  • Q. Zaheer, J. Masud Pages 385-395
    The flow field analysis of a liquid ejector pump is important for its design improvements, performance estimation and understanding of mixing and entrainment phenomenon. Ejector pumps, due to their simpler design and ease of maintenance are used in a variety of industrial applications. The subject pump, under consideration in this study, is used for transferring fuel from one fuel tank to another in a fighter aircraft. To study the underlying flow field characteristics of subject ejector pump, the fluid domain is simulated using Embedded LES turbulence modelling technique in Ansys Fluent ® environment. The flow field and performance parameters of subject pump are then compared with that of previously researched study of same pump wherein Standard K-ε RANS Turbulence Model was used. It is revealed that the results obtained using Embedded LES are much closer to experimental data than that of Standard K-ε. The limitations of RANS turbulence model for accurate simulation of complex flow field of subject pump are then identified, analyzed and discussed in details by studying the flow characteristics such as Reynolds shear stresses distribution, Potential Core estimation and turbulent viscosity modelling, obtained using both turbulent models.
    Keywords: Ejector pump, Complex flow, Reynolds shear stresses, Potential core, Embedded LES
  • N. Hajilary, A. Shahmohammadi Pages 397-404
    Gel polymer has been widely used to reduce water production in mature oil reservoirs. One of the challenges in this area is evaluation of permeability of media after the gel treatment. Darcy’s law has been used for this purpose while this equation has been developed for rigid porous media. In this study, a new mathematic model was introduced to calculate the permeability of gel coated porous media. For this purpose, a modified version of Brinkman equation was used. This model showed that permeability of gel impregnated porous media is a function of pressure drop, fluid viscosity, and gel viscoelastic properties. In order to obtain performance of new permeability model, several experiments were carried out in a porous media with radial flow. A copolymer of 2-acrylamido-2-methyl-propanesulfonic-acid sodium salt (AMPS) and acrylamide (AcA) gelant was used to form the gel in situ. Finally, to investigate the applicability of polymer gel treatment to water shut-off in porous media (sandpack), residual resistant factors (RRF) were calculated based on new permeability model.
    Keywords: Polymer gels, Brinkman equation, Radial flow, Permeability model, Darcy equation, Residual Resistant Factor
  • S. Debnath, S. Paul, A. K. Roy Pages 405-417
    The present article purposes to investigate the solute dispersion through an annular pipe in the presence of heterogeneous chemical reactions among the species and wall of the annulus. The solute is considered to experience a kinetic reversible phase exchange with the inner wall layer and irreversible absorption into the wall. Two kinds of oscillatory flow (Poiseuille and Couette flow) are considered in order to track the complex interactions between the velocity distributions and the reaction parameters. The method of moments as proposed by Aris-Barton is used to determine the apparent dispersion coefficient. The moment equations has been solved by using a standard finite difference implicit scheme, valid for small as well as large times. Dispersion coefficient due to the combined effect of reversible and irreversible reactions has been discussed in a variety of flow situations. Dispersion coefficient may be enhanced owing to the reversible and irreversible heterogeneous reactions in the boundary. On the basis of flow characteristics, radius ratio provides a mixed behaviour of the dispersion coefficient. Dimensionless mass proves to be an increasing function of reversible and irreversible boundary reaction parameters.
    Keywords: Dispersion coefficient, Reversible reaction, Irreversible reaction, Phase exchange, Damköhler number
  • T. Pant, H. Wang Pages 419-423
    An empirical model is evaluated that is in a very simple form and is often used in automobile industry to relate the pressure drop and mass flow rate in internal flows. Despite the simplicity of the model, it is remarkably accurate when it is used in a wide range of internal flows. Such accuracy and the theoretical basis of the model is not well understood, and this work aims to provide such an understanding. The theoretical basis of the empirical model is sought by performing an integral analysis based on the Navier-Stokes equation in a laminar developing channel flow. The analysis successfully yields a model that is in the same form as the empirical model. The accuracy and sensitivity of the model is then thoroughly examined through the computational studies of several internal flows. Two regimes of the model behavior in internal flows are identified, a convection dominated flow regime and a diffusion dominated flow regime. In each regime, the sensitivity of the model accuracy to the model parameters is found to be substantially different. Finally, the empirical model is applied to several more complicated internal flows to demonstrate the applicability of the model in general flows.
    Keywords: Empirical model, Internal flows, Pressure drop, Mass flow rate, CFD, Sensitivity Analysis
  • X. Mao, B. Liu, F. Yuan Pages 433-446
    Both experiments and computations are performed and analyzed to investigate the effectiveness and mechanisms of different slotted aspiration schemes in controlling the separated flows in a highly-loaded axial compressor cascade. It is found that the boundary layer aspiration on the blade suction surface can improve the incidence characteristics of the airfoil within most of the incidence range except of the extremely high incidence and the profile loss coefficient is reduced remarkably as the aspirated massflow increases. The combined aspiration is the most effective scheme to control both the separated flow on the blade suction surface and the three-dimentional hub corner separation, and an improper design of aspiration would lead to a deterioration of the flow field. Different aspiration schemes have different effectiveness in controlling the flow separation, which leads to various influences on the blade loading and the diffusion abilities. The cascade incidence characteristics of different aspiration schemes show that the part-span aspiration scheme (SS1) located on the blade suction surface can only improve the overall flow field in very high incidences, while the other schemes can reduce the overall loss coefficient within almost the whole incidence range, especially for the combined aspiration scheme. There always exists a closed separation in the cascade when the boundary layer separation is not removed completely on the blade suction surface and in the hub corner. In addition, the type of critical point is affected by the spanwise static pressure gradient, which has significant effects on the cascade performance.
    Keywords: Boundary layer separation, Aspiration, Three, dimensional corner separation, Critical point, Axial ompressor cascade
  • Q. A. Quevedo Tiznado, C. Fuentes, E. Gonz, Aacute, Lez-Sosa, C. Ch, Aacute, Vez Pages 447-457
    One of the main mechanisms of emulsion formation in porous media is the snap-off; invasion of the wetting phase flowing adjacent to the pore wall within a constriction mostly occupied by the non-wetting phase, causing breakup into isolated drops of this phase. The current approaches to determine the occurrence of this phenomenon have been formulated for quasistatic flow conditions, where the mechanisms governing the flow are controlled by the geometry of the capillary. However, some studies suggest that the drop breakup does not occur above a capillary number threshold and given a certain viscosity ratio, even if the static breakup criteria are met. In this paper, we extend the current numerical analysis of the capillary number upper limit (Calim), in which the snap-off is inhibited, by considering the effect of viscosity ratio on the dynamics of immiscible two-phase flow through constricted circular capillaries. Based on the results of this study, empirical mathematical expressions that relate the main physical variables of the flow were established as breakup criteria for dynamic flow conditions. The dynamic breakup criteria takes into account, jointly: some aspects of rheology of the two-phase system, such as the viscosity ratio; the dynamic factors of the flow, encapsulated in the local capillary number; and an integral form of the capillary geometry, represented by a parameter that relates both radii and the distance between them.
    Keywords: Drop breakup, Pore, scale flow, Local capillary number, Capillary geometry
  • T. Mothilal, K. Pitchandi, V. Velukumar, K. Parthiban Pages 459-466
    Present work optimizes the operational parameters such as solid particle diameter, inlet air velocity and inlet air temperature on heat transfer rate by Taguchi method. Operational parameters play an important role in the performance of cyclone heat exchanger thus the parameter optimization is deemed important. The parameters have been analyzed under varying solid particle diameter (300 and 400 µm), inlet air temperature (323, 373, 423 and 473 K) and inlet air velocity (5, 10, 15 and 20 m/s). Results of heat transfer rate by varying the operational parameters have been found from Computational Fluid Dynamics (CFD) software Ansys Fluent. Orthogonal array of Taguchi, the signal-to-noise ratio and analysis of variance have been employed to found the optimal parameter values and the effect of parameters on heat transfer rate. Mixed level factor and L32 array is chosen for the design of analysis in Taguchi. Result of statistical analysis shows that the developed approach yields worthy results when comparing with predicted simulation values with confidence level of 99.5%. Taguchi analysis reveals that optimized levels of parameters are 300 µm, 473 K & 20 m/s for solid particles diameter, inlet air temperature and inlet air velocity respectively. Confirmation test was conducted in simulation and experiment for optimized parameters and result shows that maximum heat transfer rate was obtained with optimized parameter among the chosen operational parameters.
    Keywords: CFD, Optimization, Taguchi method, Cyclone heat exchanger
  • I. Sreedhar, A. Sai Darshan, S. Srivastava, V. Jain Pages 467-474
    Polymer induced turbulent drag reduction has significant industrial importance and finds application in industries, oil and gas, fire-fighting, marine, irrigation, biomedical etc. Most of the reported literature is focused on the skin drag reduction in pipe flow employing drag reducing additives (DRAs) like polymers, surfactants, fibres and suspensions. In this work, the effect of polymeric addition on the total drag reduction (skin and form) is studied for turbulent flow of water through various fittings like 45 degree elbow, 90 degree miter, sudden expansion and sudden contraction. Different polymers like PAM, PEO, HPMC have been employed as DRAs at various concentrations and pressure drops. The results indicate a complex and interesting behavior. When compared to the results reported for pipe flow, even in this case polymers are found to give total drag reduction (TDR) though less relative to skin drag alone. The extent of TDR is found to depend on the nature of fitting, polymer and its concentration and the pressure drop used. From the results, it is also clear that there is a strong need to further investigate the problem using sophisticated analytical tools on rheometry and polymer degradation.
    Keywords: Form drag, Total drag, Skin drag, Polymers, Turbulent flow, Pipe fitting, Pressure drop
  • M. Ostad, R. Kamali Pages 475-481
    In the current study, the influence of blade tip clearance in different stages of a three-stage compressor is investigated. Performance diagrams of compressor were verified against experiment when there is no change in the tip clearance, after which the effect of tip clearance for the cases, 1, 1.5 and 2 mm, in the first, second and third stages of rotor was studied. The results indicated that the impact of tip clearance increase did not have any effect on choked flow rate value in the first and second stages, and only the change in the third stage tip clearance reduced the choked flow rate. For the same tip clearance value, the highest compressor performance loss occurs in the case of applying the tip clearance in the third stage, which is also the final stage and is highly sensitive to tip clearance changes. Moreover, modifying the tip clearance is effective on the flow angle in the trailing edge and as the tip clearance increases, the same thing happens about the flow angle. The maximum value of flow angle changes by modifying the tip clearance belongs to the third stage of the compressor.
    Keywords: Variable tip clearance, Stall margin, Flow angle, Compressor performance
  • J. D. Quadros, S. A. Khan, A. J. Antony Pages 483-496
    Effectiveness of active control of micro jets has been examined by conducting experiments through an abruptly expanded axi-symmetric duct in a view to control base pressure. For this purpose, 1mm orifice diameter micro jets have been deployed at an interval of 900 along the exit diameter of the nozzle. The experiments have been conducted by considering three flow parameters at three levels. Mach number (M), length to diameter (L/D) ratio and area ratio (AR) are the three parameters used to conduct and analyze the flow experiments. Base pressure is considered to be the response variable. The experimentation has been carried out for two cases, i) without active control; ii) with active control. An L9 orthogonal array has been implemented to plan the experiments. It is observed that the control becomes effective for lower area ratios when compared to the higher ones. In addition to this, at high area ratios suction at the base decreases and hence base pressure continuous to diminish with increasing L/D until it reaches a value of L/D=6. The obtained experimental results are subjected to multiple linear regression analysis and Analysis of variance (ANOVA). The performances of the two linear regression models were tested for their prediction accuracy with the help of 15 random test cases. It is observed that, both linear regression models for base pressure without and with control are statistically adequate and capable of making accurate predictions. Furthermore, this work also concludes that, Mach number is the most significant factor affecting base pressure followed by area ratio and L/D ratio for both cases of experimentation. The obtained experimental results are further validated by CFD analysis and are found to be in good concurrence with each other.
    Keywords: Base pressure, Mach number, Area ratio, Length to diameter ratio, Analysis of variance
  • S. Kumari, P. V. S. N. Murthy Pages 497-505
    The instability of non-Newtonian power law fluid in double diffusive convection in a porous medium with vertical throughflow is investigated. The lower and upper boundaries are taken to be permeable, isothermal and isosolutal. For vertical throughflow the linear stability of flow is determined by the power law index (n), non-Newtonian Rayleigh number (Ra), Buoyancy ratio (N), Péclet number (Pe) and Lewis number (Le). The eigenvalue problem is solved by two-term Galerkin approximation to obtain the critical value of Rayleigh number and neutral stability curves. It is observed that the neutral stability curves, as well as the critical wave number and Rayleigh number, are affected by the parameters such as Péclet number, buoyancy ratio and Lewis number. The neutral stability curves indicate that power law index n has destabilizing nature when it takes values for dilatant fluid at low Péclet numbers while for the pseudoplastic fluids it shows stabilizing effect. In the absence of buoyancy ratio and vertical throughflow, the present numerical results coincide with the solution of standard Horton-Rogers-Lapwood Problem. The numerical analysis of linear stability for the limiting case of absolute pseudoplasticity is also done by using Galerkin method.
    Keywords: Porous medium, Non, newtonian fluid, Buoyancy ratio, Rayleigh number, Lewis number
  • M. Whalley, B. Skews Pages 507-517
    The flow field which results from an expansion wave entering a cavity from an upstream tube, and the focusing effect which occurs, is investigated. Different cavity geometries, different expansion wave pressure ratios and different expansion wave widths are explored. As the expansion wave propagates into the cavity it induces flow in the opposite direction and back down the walls. The flow experiences compression as it flows out back into the tube because of the concave surface of the cavity it encounters. This can result in the formation of shock waves which can propagate back up into the cavity. Very low pressure and temperature regions can develop because of the focusing action of the expansion. A convenient way of generating an expansion wave numerically and/or experimentally is in a shock tube. This consists of a tube divided into two compartments, one at high pressure and one at low pressure separated by a frangible diaphragm. On bursting the diaphragm, a shock wave travels in one direction and an expansion in the other towards the cavity. Whilst ideal boundary conditions can be imposed in numerical simulation laboratory experiments are complicated by the diaphragm being curved and having a finite opening time. The effect of an initially curved diaphragm is briefly considered. The expansion wave pressure ratio was altered by changing the initial pressure ratio across the diaphragm. For an initially high pressure ratio, supersonic flow can occur behind the trailing edge of the expansion wave which has a marked influence on the flow. The width of the wave is dependent on the distance of the diaphragm from the cavity and also has a significant influence on the flow. As the width of the wave increases and the density gradient decreases, focusing effects becomes significantly weaker. Correspondence between experiment and simulation is examined.
    Keywords: Compressible flow, Unsteady flow, Wave focusing
  • S. Huang, Y. Xu, L. Zhang, W. Zhu Pages 519-526
    Domain decomposition is involved in Fluid-Structure Interaction (FSI) analysis to speed up their computations. Non-matched meshes always exist in the interface of these different domains which brings data exchange problem. A load transfer method is investigated in this article to deal with non-matching meshes between fluid and structure. The local nearest neighbor searching algorithm was used in this method to match fluid nodes and structural elements, while thin plate splines with tension were used to deal with data transfer between non-matching meshes in FSI computations, and the corresponding matrix equations for the target points are presented. Implementations of the obtained algorithms were used to solve the one-way FSI problem of the CRH380C high-speed train and the relative error of transferred results was analyzed. The statistical parameters under two algorithms, the TPS model and the model combining both TPS model and nearest interpolation model were compared and the results indicate that the latter can transfer data more accurately.
    Keywords: Load transfer method, High, speed train, Data exchange, Interpolation method, Thin plate spline with tension