EI、Scopus 收录

2007 Vol. 23, No. 3

Display Method:
Effect of vectoring jet on asymmetrical vortex of slender body
Yankui Wang, Yongsheng Zhang, Xueying Deng, Shuifeng Yang, Xiaowei Yu
This paper studies the effect ofvectoring jet on the transverse aerodynamic characteristics of a slender body. Theresults show that: 1) the side force of the slender body caused byasymmetric vortices at a high attack angle can be reduced by vectoring jeteffectively. 2) with the increase of the attack angle, the affected zone of vectoringjet is expanded and reaches the nose of the slender body. 3) with theincrease of thepressure of the vectoring jet, the effect of jet on the side forceof the slender body is increased. The Mach number of the nozzle is thekey simulation criterion for the thrust vectoring test. The test wascarried out in D4 wind tunnel in Beijing University of Aeronautics andAstronautics.
2007, 23(3): 289-296. doi: 10.6052/0459-1879-2007-3-2006-140
Hydrodynamic characteristics of a vertical thin barrier in a continuously stratified fluid
Yunxiaing You, Jie Xu, Gang Wei, Dongqiang Lu
This paper discusses the interaction of water waves with a verticalthin barrier in a continuously stratified fluid. Under the Boussinesqapproximation, the dispersion relations are obtained for planemonochromatic waves in a continuously stratified fluid, and the theoreticalsolutions are obtained by the eigenfunction expansion for the reflectionand transmission energies, as well as the exciting forces on the barrier dueto incident waves. For a fixed frequency $\omega $, when $\omega >N$($N$ being the buoyancy frequency), there is only a monochromatic wavepattern with one mode. When $\omega < N$, there are monochromatic wavepatterns with infinite modes, and it is shown that for the incident wave ofeach mode, the energies of the reflection and transmission waves of anyother mode are equal. The reflection and transmission energies, as well asthe exciting forces on the barrier due to incident waves are computed fortwo types of structures: a surface-piercing barrier and abottom-touched one, respectively. The results indicate that the densitystratification can have a significant effect on the hydrodynamic characteristicsof the barrier over a wide range of frequencies. In particular, when $\omega< N$, the transferred energies and the exciting forces on the barrier due toincident waves of the first mode are much less than those due to incidentwaves of other modes.
2007, 23(3): 297-310. doi: 10.6052/0459-1879-2007-3-2005-196
Direct numerical simulation of turbulent pipe flow controlled by MHD for drag reduction
Shidong Luo, Chunxiao Xu, Guixiang Cui
We studied the MHD control of drag reduction in pipe flow through DNS. TheLorentz force is generated in the conductivity fluid by an array ofelectrodes and magnets fixed on the inner face of the pipe, when the voltageis applied on the electrodes. The drag is decreased numerously by changingthe amplitude and the period of the force. We studied the relation betweenthe ratio of drag reduction and the control parameters, and the physicalmechanism is discussed additionally. The MHD control possesses the optimumperiod, the maximum drag reduction is 24.2{\%}; The MHD control depressesthe fluctuations of the streamwise and normal velocity, but the azimuthalvelocity fluctuations grows with the increasing of the oscillatory period.
2007, 23(3): 311-319. doi: 10.6052/0459-1879-2007-3-2006-296
EMD simulation study on potential and force at liquid-vapor interface system composed of water molecules
Jie Sun, Yaling He, Yinshi Li, Wenquan Tao
The interaction of potentials between molecules is important for the molecular interface behavior. Theliquid-vapor interface system composed of water molecules is simulated andstudied with equilibrium molecular dynamics (EMD) simulation method in this paper.The results agree with the known conclusions that there is a marked fall ofpotential between the liquid bulk and the vapor bulk. It is also indicatedthat different types of forces have different effects on the moleculesmoving across the interface. The L-J force acts as a resistance tocondensation while the electrostatic force, which is predominant in theresultant force, acts as motivity. Meanwhile, some typical molecules whichparticipate in the phase-change process, are especially traced. The resultsshow that the process of condensation leads to a drop of thepotential energy of a vapor molecule, which falls into the potential well.Correspondingly, the process of evaporation increases thepotential energy of a liquid molecule, which gets over the potentialbarrier.
2007, 23(3): 320-324. doi: 10.6052/0459-1879-2007-3-2006-076
Wave propagation on non-uniform currents and depth
Hongsheng Zhang, Hongjun Zhao, Pingxing Ding, Guoping Miao
By transforming two time dependent hyperbolic mildslope equations with dissipation term for wave propagation on non-uniformcurrents into equivalent equations, respectively, the effects ofdissipation onintrinsic frequency and wave number are analyzed to choose the suitablemathematical model, in which the wave number vector and intrinsicfrequency are expressed both more rigorously and completely. By using theperturbation method, a time dependent parabolic equation is obtained fromthe time dependent hyperbolic mild slope equation for asuitable mathematical model, and solved by using the alternating directionimplicit method. A numerical model is built for wave propagation andtransformation on non-uniform currents in water of slowly varyingtopography. Comparisons are made between the numerical solutions and thetheoretical solutions for the case of collinear waves and current, anda good agreement is found. Based on the interactions between incident wave and currenton a sloping beach (Arthur,1950), the differences of wave number vectorbetween refraction and combined refraction-diffraction of waves arediscussed quantitatively, while the effects of different methods ofcalculating wave number vector on the numerical results are shown.
2007, 23(3): 325-332. doi: 10.6052/0459-1879-2007-3-2006-006
Experimental research on flow induced oscillation of circle moonpool
Xiongliang Yao, Zhuang Kang
Moonpool is just like a cave-in cavity with a free surface.With the help of empirical mode decomposition (EMD) method this experiment wasconducted in a towing tank. The key part of the EMD method is that complicated data set can be decomposed into a finite and oftensmall number of `intrinsic mode functions (IMF)'. By a comparative method, thedata were collected with and without the moonpool in all conditionsseparately. With the empirical mode decomposition, theintrinsic mode functions of the data, and then the PSDspectrum of each IMF are obtained. From the difference of each PSD spectrum betweenthe results with and without moonpool, the data for the flowover the moonpool can be obtained.For the moonpool, the fluctuating force consists of the water oscillation andthe shear layer self sustained oscillation. The frequency of the wateroscillation is determined by the shape and size of the moonpool. Thefrequcney of the shear layer self sustained oscillation is also related withthe flow velocity over the moonpool, which will change with thevelocity. When the two frequencies are close to each other, a resonance willhappen. In this experiment, when the Froude number falls within therange of 0.26 to 0.39, theresonance phenomenon is observed. It appears that the water in themoonpool moves upand down in the vertical direction. This is so-called the ``Piston'' phenomenon.When the Froude number is greater than 0.52, there is no resonance phenomenonany more. In this paper, under some assumptions and the potential flow theory,the resonance frequency of the water in the moonpool is calculated,which is very close to the experiment result.In a resonance, the fluid fluctuating force in the moonpool isa harmonic force. The frequency is the same as the resonance frequency andthe amplitude is great. When there is no resonance, the fluctuating force can be obtained with the EMD method, which consists of the shear layer oscillation and the water oscillation.The wall pressure is also measured in this investigation. The wall pressureof the moonpool is found quite different not only in value but also in distribution in the circumferential direction and the vertical direction,whether in resonance or not.
2007, 23(3): 333-342. doi: 10.6052/0459-1879-2007-3-2005-443
Numerical study on backward-forward double-detonation driver for high enthalpy shock tubes
Jinping Li, Zonglin Jiang, Hong Chen, Heng Feng, Hongru Yu
A numerical investigation on the performance of abackward-forward double-detonation driver for high enthalpy shock tubes wasconducted by solving one-dimensional Euler equations with the DispersionControlled Dissipative (DCD) scheme. The modified two-step chemical reactionmodel was applied to describe chemical reactions of the oxygen-hydrogenmixture. The numerical results indicate that if the initial pressure ratioof the auxiliary driver section to the primary driver section is less thanthe critical value as was estimated under the condition of no Taylorexpansion waves in the driver, Taylor expansion wave is still observablewith a smaller angle of the expansion wave fan; if the initial pressure ratiois set to be equal to the critical value, Taylor expansion waves disappearand the incident shock wave does not attenuate in the driven section; if theinitial pressure ratio is taken to be larger than the critical value, notonly Taylor expansion waves disappear, but also an over-driven detonationwave develops in the detonation driver and can increase further the drivingpressure.
2007, 23(3): 343-349. doi: 10.6052/0459-1879-2007-3-2005-432
Development of macro-depth-sensing-indentation instrumentation
Dongxu Liu, Taihua Zhang, Yong Huan
Using an Instron 5848 Microtester assembled with an additional displacementsensor and specially designed jigs as the main loading setup,DSI(depth-sensing indentation) tests were performed on typical metallicmaterials at macro range. From the indentation data, the hardness andmodulus of the materials were directly extrapolated without frame compliancecalibration employing an analysis procedure proposed by Oliver and Pharr. Inthis analysis the contact area was calculated for indentation depths \ge6\mu m from the area function of an ideal indenter tip and for depths <6\mu m from the area function suggested by Liu and Zhang. A comparisonbetween the our results and those from the MTS NanoIndenter XP showed that the repeatability isideal, and the scatter of both hardness and modulus values is within 10%.So, it is feasible to develop the DSI capacity of traditional materialtesting machines.
2007, 23(3): 350-355. doi: 10.6052/0459-1879-2007-3-2006-060
The energy equation adjusting algorithm for the nonlinear dynamic equation with multi-degree of freedom
Weidong Li, Hexiang Lv
The energy adjust numerical algorithm for the nonlineardynamic equation with multi-degree of freedom has been proposed. First, bythe Taylor expanded formula and Duhamel integration, an integral iterationformula which has an undetermined parameter has been achieved. Second,through some mathematic approach, the original dynamical equation has beenconverted to an energy adjusting equation which can determine theundetermined parameter. Finally, substituting the parameter into theintegral iteration formula, an accurate numerical value has been achieved.All examples show that the method can eliminate the algorithm damping andhas better stability than the Runge-Kutta method under bigger integral step.
2007, 23(3): 356-364. doi: 10.6052/0459-1879-2007-3-2006-280
Scattering of plane rayleigh waves by a circular-arc alluvial valley: an analytical solution
Chenggang Zhao, Lei Wang, Fuping Gao
The scattering and diffraction of plane Rayleigh waves in acircular-arc layered alluvial valley with soft-soil deposits is studiedanalytically in this paper. The soft sediments in the circular-arc valleyare simulated by applying Biot's dynamic theory to a saturated porous medium, andthe half-space is assumed as an elastic single-phase medium. In this analysis,Fourier-Bessel series expansion technique is employed to obtain theanalytical solutions. Based on the analytical solutions, the effects ofdimensionless frequency, the layered sediments sequence and the relativerigidity and thickness on the scattering and diffraction are investigated.
2007, 23(3): 365-373. doi: 10.6052/0459-1879-2007-3-2006-145
Precise integration method for duhamel terms arising from non-homogenous dynamic systems
Shujun Tan, Wanxie Zhong
With the precise integration method (PIM) proposed for lineartime-invariant systems, one can obtain precise numerical results approaching theexact solution at the integration points. However, it is more or lessdifficult to use the algorithm in the Duhamel's integration arisingfrom the non-homogenous dynamic systems due to the inverse matrixcalculations. So the precise integration method for Duhamel terms withoutthe inverse matrix calculations is proposed. By applying the techniques ofaddition theorem and increment storage, which are the key ideas of PIM,directly to the Duhamel integration terms, it can also give precisenumerical results can be obtained close to the computer precision when the non-homogenous termsare polynomial, sinusoidal, exponential or theircombinations. In particular, this method is not affected by the qualityof the system matrix (or the relative matrix). If the system matrix issingular or nearly singular, the advantages of the method will be more remarkable.Numerical examples are given to demonstrate the validity and efficiency ofthe method.
2007, 23(3): 374-381. doi: 10.6052/0459-1879-2007-3-2006-553
A generalized maugis model for adhesive contact of arbitrary axisymmetric elastic objects
Zhi-Jun Zheng, Jilin Yu
Based on a linear combination of the solutions derived fromthe Sneddon method and the Lowengrub-Sneddon method, a general solution ofthe axisymmetric problem is obtained for the elastic half-space with mixedboundary conditions. And then, the frictionless and adhesive contact problembetween two general axisymmetric elastic objects is studied. For anarbitrary effective surface profile, i.e., the initial contact occurs at thecentral part, and an arbitrary surface adhesive interaction, a generalizedMaugis model is derived and it can be divided into two parts correspondingto the contributions of the surface profile and the surface adhesiveinteraction, respectively, and a coupling relation between the deformationand the adhesive interaction. Based on the Dugdale model for the surfaceadhesive interaction, a generalized M-D model is derived for an arbitraryeffective surface profile. Two extremes are found for this model. For ashort-range strong interaction or compliant material, its limiting form iscorresponding to the generalized JKR model. And for a long-range weakinteraction or stiff material, another limiting form is corresponding to ageneralized DMT model.
2007, 23(3): 382-388. doi: 10.6052/0459-1879-2007-3-2006-395
Multiobjective optimization design of aseismic steel frames using genetic algorithm
Jizhuo Huang, Zhan Wang
Considering the fact that the optimal design of an aseismicsteel frame is a multiobjective optimization problem, an optimization methodfor the multiobjective optimization design of aseismic steel frames based ongenetic algorithm (GA) was explored. A new ranking approach without usingpenalty function methods was presented to handle a constrainedmultiobjective optimization problem. This approach can deal with objectiveand constraint functions respectively. Based on the new ranking approach, aGA-based optimization method for constrained multiobjective optimizationproblems (CMOPGA) was advanced, and its flow chart was given. To minimizethe weight of an aseismic steel frame and its total dynamic strain energy, amathematical formulation of the multiobjective optimization design of anaseismic steel frame was established based on some corresponding codes. Anexample of two-bay six-story aseismic steel frame was provided, and acompromise principle of relative minimum distance was proposed for designersto select the compromise solution from a Pareto optimal set in the absenceof engineering experience. The optimal results show that it is effective forthe multiobjective optimization design of aseismic steel frames by usingCMOPGA.
2007, 23(3): 389-397. doi: 10.6052/0459-1879-2007-3-2006-373
Nonlinear numerical analysis of progressive damage of 3D braided composites
Kun Xu, Xiwu Xu
A tensile strength model to predict the damage initiationand propagation of 3D braided composites has been proposed based on therepresentative volume element by using the nonlinear finite element method.The geometrical unit-cell model considering the mutual contact and jammingof yarns was applied in the FEM model coupled with the periodicaldisplacement boundary conditions. In particularly, the nonlinear shearstress-strain behaviors of the yarns were considered in the model andHashin-type failure criteria were adopted to judge the occurring of damageevents of the yarns. Based on the degradation scheme of elastic mechanicalproperties, the simulation of the whole process of damage initiation,propagation and catastrophic failure of the structure was done in detail. Toverify the numerical model, two typical specimens with different braidingangle were selected to conduct the simulations. The calculated values by themodel are well agreed with the measured values. Numerical results verify theeffectiveness of the model. The effects of the shear nonlinearity of theyarn and damage accumulation on the nonlinear mechanical behavior of 3Dbraided composites were discussed. The conclusion has indicated that thenonlinear mechanical behaviors of the composites are affected remarkably bythe shear nonlinearity of yarns and accumulation of damages when thebraiding angle of the composites is increased.
2007, 23(3): 398-407. doi: 10.6052/0459-1879-2007-3-2006-450
Analysis of vortex shedding modes of an in-line oscillating circular cylinder in uniform flow
Yufei Zhang, Zhixiang Xiao, Song Fu
Flows around stationary and in-line oscillating circularcylinder in uniform flow at Reynolds number of 150 is studied by solvingNavier-Stokes equations on a moving grid system which described by ArbitraryLargangian-Eulerian (ALE) method. The main objective is to investigate thelift and drag features and vortex shedding modes through different reducedoscillating frequencies and a little strong amplitudes, where the ratio ofamplitude and diameter A/D is from 0.3 to 1.2. Present numerical resultsshow that there are several modes of vortex shedding such as symmetrical Sand anti-symmetrical A-I, A-III and A-IV modes. Through analysis of previousand present results, the region of lock-on is extended, which is dividedinto five sub-regions. Especially, the force on the circular cylinder atanti-symmetrical A-I mode shows more complex characters than the originalmode. Based on the experimental and numerical results, we suggest that theradio of maximum acceleration Af_{c}^{2}/Df_{s0}^{2} should be animportant parameter for vortex shedding modes, especially for thesymmetrical S mode.
2007, 23(3): 408-416. doi: 10.6052/0459-1879-2007-3-2005-515
Evolution of the fluid mixing zone in richtmyer-meshkov instability at a gas/liquid interface
Honghui Shi, Qiwei Zhuo
This paper presents an experimental study on fluid mixinginduced by Richtmyer-Meshkov (RM) instability at an interface between waterand air using a rectangular shock tube. The length of mixing zone is foundto obey a linear law $h \propto t$ when the Atwood number approaches to 1under different Mach numbers and initial perturbations. However, the lengthof mixing zone increases when Mach number increases from 1.45 to 1.7, alsoincreases when initial perturbations are random., and the mixing goes moretempestuously.
2007, 23(3): 417-421. doi: 10.6052/0459-1879-2007-3-2006-055
A contact algorithm for parallel computation of FEM
Fujun Wang, Liping Wang, Jiangang Cheng, Zhenhan Yao
A general contact algorithm for parallel finite element simulation ofdynamic contact problem is presented in this paper. The local Lagrangemultiplier approach is introduced into the new contact algorithm. Since theimpenetrability condition and the interaction of adjacent contact pair areall considered in this algorithm, the contact constraint and equilibriumequations of the system are satisfied sufficiently as compared withtraditional penalty method. Although some local iterations have to be done,the algorithm has high computational efficiency, and is totally consistentwith explicit time integration. In addition, the existing contact searchalgorithms developed for serial finite element program could be transferedinto the parallel finite element program without any modification bywonderful design of a special domain decomposition scheme. Numericalexamples show that the new contact algorithm is suitable for parallel finiteelement computation. The new algorithm ensures not only the good simulationaccuracy, but also the high parallel-execution efficiency.
2007, 23(3): 422-427. doi: 10.6052/0459-1879-2007-3-2006-407
Source of attachment forces of ants and the secretion effect
Qun Zhou, Bin He, Jiguang Yue
A test platform was developed based on centrifugaltechnology. Attachment forces of ants were measured to be 40 times of thebody weight. Configuration of claws and pad of ants were observed usingscanning electron microscope. Analysis results suggest that the secretion isvery important. Excluding vacuum force, electrostatic force and othermechanisms, it is likely that capillary force is more significant in thevertical direction and viscosity of the adhesive liquid film dominates inhorizontal pulls. Analysis using ANSYS indicates microfolds on the padsurface facilitate the drainage of the liquid. These results promoteunclosing adhesion mechanisms of insects.
2007, 23(3): 428-432. doi: 10.6052/0459-1879-2007-3-2006-189