Table of Content

    18 January 2010, Volume 42 Issue 1
    Research paper
    Model of inflatable structure/fluid interaction for variable leading edge
    Yuewen Jiang Zhengyin Ye Zhengke Zhang
    2010, 42(1):  1-7.  DOI: 10.6052/0459-1879-2010-1-2008-386
    Asbtract ( 1269 )   PDF (1161KB) ( 594 )  
    Related Articles | Metrics
    The paper presents the interactive motion equation of inflatable structure in flow. The inflatable structure is used in the technique of a variable leading edge for high lift. The interaction between structure and fluid is a typically aeroelastic problem. The motion equation is obtained through the analytic solution of an infinitesimal element in the discrete membrane. Coupled with Navier--Stokes equation, the aeroelastic equation was solved with time-marching methods, and the aerodynamic characteristics of NACA 63-212 airfoil with the inflatable leading edge (ILE) were obtained. From the results, the ILE technique has been applied to significantly increase the stall angle of attack and lift coefficient. The values of stall angle of attack and maximum lift coefficient were enhanced to be 30 percent and 22 percent, respectively. In addition, the deformation of ILE greatly influenced the flow around the airfoil while the elastic coefficient of material was low. The inflatable structure near the suck peak deformed clearly, and this hump could induce the separation in advance.
    Research on statistic characteristic of near-wall turbulence
    Honglin Xiao Jisheng Luo
    2010, 42(1):  8-14.  DOI: 10.6052/0459-1879-2010-1-2008-593
    Asbtract ( 1590 )   PDF (962KB) ( 526 )  
    Related Articles | Metrics
    Channel turbulence is simulated by large eddy simulation (LES), and turbulent databases of channel flows at different Reynolds numbers are established. Then, the mean velocity distribution, Reynolds stress and root-mean-square (RMS) velocity distribution are obtained. Furthermore the problem of the wall law is discussed. Meanwhile, a rule of Reynolds stress, energy of velocity fluctuations and RMS velocity with different Reynolds number is presented along with quantitative formula.
    Numerical studies on flows around a flexiblly wiggling hydrofiol
    Guoyi He Xing Zhang Shuguang Zhang
    2010, 42(1):  15-22.  DOI: 10.6052/0459-1879-2010-1-2008-570
    Asbtract ( 1423 )   PDF (1435KB) ( 695 )  
    Related Articles | Metrics

    Fishes often use a wiggling motion to generate propulsion for swimming. The wiggling motion can be modeled by a progressive wave. In the present study, an immersed boundary method is used to simulate the flows around the wiggling hydrofoil NACA 65-010 at low Reynolds numbers, and the effects of Reynolds numbers on the propulsion generation are investigated. It is observed that, as the Reynolds number ranges from 1 to 200, both the thrust coefficient and the propulsion efficiency increase with the increase of the Reynolds number, while the power efficiency decreases. Especially in the ranges of the Reynolds number smaller than 20, the lift, thrust and power coefficients exhibit a large change. With the increase of the Reynolds number, the flow patterns around the wiggling foils become more complex: the vorticity becomes more intensive near the trailing edge and the wiggling foil produces reversed von Karman vortex streets to generate propulsion.

    The nonlinear water wave equations with full dispersion
    Jin Hong Zhili Zou
    2010, 42(1):  23-34.  DOI: 10.6052/0459-1879-2010-1-2008-080
    Asbtract ( 1296 )   PDF (1740KB) ( 436 )  
    Related Articles | Metrics
    A 2D nonlinear water wave model with full dispersion is developed. The model is based on the nonlinear kinematic and dynamic free surface boundary conditions and is expressed in terms of free surface elevation $\eta $ and the velocity potential $\phi _\eta $ at the free surface. The derivation of the equations is accurate to third order in nonlinearity and keeps exact dispersion. The mild slope assumption is adopted and the derived equations can be seen as the extention of the mild slope equation of Berkhoff (1972) to the nonlinear and irregular wave case. The corresponding numerical scheme is presented, and the special attention is paid on the treatment of the integration terms in the equations. The validation of the model is made by simulating the first and second order Stokes waves and the nonlinear evolution of wave groups, the advantage of the model is shown by the good prediction of amplitude dispersion and four-wave resonant interaction for the wave group evolution.
    Experimental study of crescent wave without wind driving
    Yalong Zhou Zhili Zou
    2010, 42(1):  35-43.  DOI: 10.6052/0459-1879-2010-1-2008-143
    Asbtract ( 1223 )   PDF (2222KB) ( 479 )  
    Related Articles | Metrics
    Experiment study on crescent waves was performed without wind driving. L2-type crescent waves wereobtained in the experiment, as well as other two new types of three-dimensional wave patterns. The time series and amplitude spectra characteristics of generated crescent waves were analysized, and wave lengths of transverse waves and wave heights of L2-type crescent waves were discussed.
    Dynamic interaction between waves and poroelastic seabed with boundary layer effects
    2010, 42(1):  44-50.  DOI: 10.6052/0459-1879-2010-1-2008-227
    Asbtract ( 1298 )   PDF (1112KB) ( 551 )  
    Related Articles | Metrics
    Dynamic interaction between waves and poroelastic seabed is investigated in relation to the boundary layer near the seabed. Potential theory is applied to describe the wave field and a boundary layer modification is considered. The deformation of the seabed is governed by Biot's consolidation theory. Stress and velocity continuity at the interface will be coupled with the wave motion and the relevant seabed response. Under the assumptions of small amplitude waves and small deformation of the seabed, an analytical solution is obtained and shown to be in good agreement with field data. The conditions under which wave-seabed interaction becomes significant are also discussed.
    Leaky rayleigh wave at the interface between inviscid fluid and solid
    Xiao Chen
    2010, 42(1):  51-55.  DOI: 10.6052/0459-1879-2010-1-2007-563
    Asbtract ( 1324 )   PDF (806KB) ( 607 )  
    Related Articles | Metrics
    The leaky Rayleigh wave, an inhomogeneous surface wave, exists at the interface between a fluid half space and a solid half space. The secular equation of leaky Rayleigh wave is derived and the formulas of particle displacement and stress distributions in the inviscid fluid and the isotropic solid are presented. The characteristics of leaky Rayleigh waves, the particle displacement and stress distributions in the fluid and solid are investigated with numerical results using those equations. The effect of the fluid-solid density ratio on the velocity and attenuation of leaky Rayleigh wave and its particle displacement and stress distributions in the fluid and the solid is analyzed numerically. The leaky Rayleigh wave is non-dispersive and attenuate with propagetion distance. Its velocity and attenuation increase with the fluid-solid density ratio. The fluid-solid density ratio has significant effect on the vertical displacement but little on horizontal displacement in the solid half space. The fluid-solid density ratio also dramatically effects on the vertical normal stress and shear stress but little on horizontal normal stress in the solid half space. The theory is promising to be explored to nondestructive testing and evaluation with leaky Rayleigh wave.
    he analytical solutions of thick-walled cylinder of softening material and its stability
    Youquan Yin
    2010, 42(1):  56-64.  DOI: 10.6052/0459-1879-2010-1-2008-440
    Asbtract ( 1445 )   PDF (1259KB) ( 499 )  
    Related Articles | Metrics
    The constitutive law of elastoplastic material being simplified to three-line model (elastic-linear softening plastic-residual ideal plastic model), and the material obeying Tresca yield criteria and associated flow rule, the analytical solutions of thick-walled cylinder subject to internal pressure $p$ were derived in the paper. The result shows that the yield stress in the softening plastic region is the inverse square of radial coordinate $r$. Firstly, the pressure $p$ was taken as generalized force, the displacement $u$ taken as generalized displacement, and the thick-walled cylinder taken as a whole system. On the basis of the solutions the stability problem of thick-walled cylinder was then discussed. The $p$-$u$ curve of balance path was drawn, on which each point denotes a balance state. The slope of the tangent line for each point can be considered the stiffness of thick-walled cylinder. The extreme value of generalized force is the critical point on the curve, and the critical point separates the curve into two sections: the section before the critical point is stable, and the stiffness is positive; the section after the point is unstable, and the stiffness is negative. When the generalized force reaches the critical point, the displacement increases quickly and the system loses its stability, while ideal plastic thick-walled cylinder loses its stability only when the plastic region penetrates through the whole cylinder. Therefore, the failure mechanism is completely different: the former belongs to extreme value point destabilization, and the latter belongs to strength failure. That is to say, the bearing capacity has different mechanical meanings.
    Equivalences of the generalized viscoelastic models and their inherent properties
    Xiaohua Huang Xiating Feng Bingrui Chen
    2010, 42(1):  65-73.  DOI: 10.6052/0459-1879-2010-1-2008-128
    Asbtract ( 1337 )   PDF (1346KB) ( 567 )  
    Related Articles | Metrics
    The equivalences of generalized viscoelastic models and their inherent properties are investigated. The creep compliance of the generalized Maxwell model is difficult to obtain due to its parallel connection. However, a method by using Laplace transform is proposed in the present paper to solve this problem and a theoretical expression is obtained. The equivalence of the generalized Maxwell model and the Kelvin chain is proved and the conversion relations of the parameters between these two models are also derived. Moreover, an inherent property of the generalized viscoelastic model proved here is that relaxation times and delay times are mutually interlaced, and the former is always greater than the latter when the spectrums of relaxation time and delay time are placed in order. This property is also the imperative condition when the generalized viscoelastic model is applied to describe the rheological behavior of materials in engineering. As special instances of the equivalence of the generalized Maxwell model and the Kelvin chain, there are the equivalence between the Wiechert model and the generalized Zhong-cun model, the Generalized Jeffreys model and the Generalized N-K model, the Maxwell chain and the Generalized Kelvin model. Finally, the proposed method is demonstrated by an example.
    discontinuous galerkin method for discontinuous temperature field problems
    Donghuan Liu Xiaoping Zheng Yinghua Liu
    2010, 42(1):  74-82.  DOI: 10.6052/0459-1879-2010-1-2009-003
    Asbtract ( 1539 )   PDF (1456KB) ( 747 )  
    Related Articles | Metrics
    A discontinuous Galerkin (DG) finite element method for the discontinuous temperature field problems is presented. The DG method uses discontinuous interpolation functions on the element boundaries, and the discontinuous effect is considered by the penalty function techniques, in which the numerical flux and the stabilization term are adopted at the interface. By substituting the numerical flux at the imperfect contact interface with the definition of the thermal contact resistance, and eliminating the stabilization term, the present DG method can easily and accurately capture the temperature jump caused by thermal contact resistance. Compared with the continuous Galerkin method, the present DG method also has higher computational efficiency in capturing the peak value of the heat flux of the local high gradient temperature field. Numerical examples also show that the present DG method is a novel numerical method for solving the coupling problems between the temperature and stress field caused by thermal contact resistance.
    Area coordinates and b-net method for quadrilateral spline elements
    Juan Chen Chongjun Li Wanji Chen
    2010, 42(1):  83-92.  DOI: 10.6052/0459-1879-2010-1-2008-501
    Asbtract ( 1114 )   PDF (1415KB) ( 523 )  
    Related Articles | Metrics
    In general, there are two types of quadrilateral isoparametric elements, Serendipity type and Lagrangian type. The S-type elements only possess low order completeness, and are sensitive to mesh distortions. The L-type elements possess high order completeness, but include interior nodes. By using numerical integrations due to isoparametric transformation, the overall stiffness matrix may remain singular. In this paper, a kind of quadrilateral spline elements are constructed by using triangular area coordinates interpolation and B-net method. These spline elements have property of conformality, and are insensitive to mesh distortions. The 8 and 12-node quadrilateral elements are represented by bivariate splines of degree 2 and 3, respectively. The two elements possess 2 and 3 order completeness in Cartesian coordinates, higher than the corresponding isoparametric elements with the same nodes. Some numerical examples are employed to evaluate the performance of the proposed elements. The results show that the new spline elements present higher precision and efficiency in comparison with other quadrilateral elements.
    Influence of the stress path on dilatancy of soils and its modeling
    Ding Luo Zhiwei Gao Yangping Yao
    2010, 42(1):  93-101.  DOI: 10.6052/0459-1879-2010-1-2008-484
    Asbtract ( 1159 )   PDF (1196KB) ( 1065 )  
    Related Articles | Metrics
    One of the most important characteristics of soil is its dilatancy in which plastic volumetric strain occurs during shearing. The influence of stress path on the dilatancy of soils is systematically analyzed in this paper with the asymptote state model proposed by authors et al (Luo et al, 2009), in which the part coupling effect between the plastic volumetric strain and plastic shear strain is assumed firstly. The comparisons are carried out between the test results of the dilatancy and the model predictions of the original Cam-Clay model, the modified Cam-Clay model, the unified hardening model (Yao et al, 2008) and our proposed model. The comparisons results demonstrate that the asymptote state model can describe the effect of the stress path on the dilatancy of soils in a better way in the stress-strain and strength behaviors of soils.
    The biomechanical study of the subsequent injury of an acl deficient knee
    Jie Yao Yubo Fan Ming Zhang Deyu Li He Gong
    2010, 42(1):  102-108.  DOI: 10.6052/0459-1879-2010-1-2008-526
    Asbtract ( 1252 )   PDF (1481KB) ( 489 )  
    Related Articles | Metrics
    Meniscus and ligament ruptures are common in anterior cruciate ligament (ACL) deficient knee. The mechanism of injuries is still unclear, and was mainly inferred from ex vivo experiments and clinical studies. In this study, a three dimensional finite element model of knee joint was reconstructed from magnetic resonance images. The geometry data of the knee from literature was used to validate the model. Tibial displacement and in situ force in ACL in response to the anterior tibial loads were calculated and coincided with the published experimental data. The simulations of the models with and without ACL were performed under anterior tibial loads and axial tibial loads at 0$^\circ$ and 30$^\circ$ flexion. Our results suggest that ACL deficiency could lead to abnormal stress distribution: the stress near the posterior insertion of the medial meniscus increased significantly; the changes of the stress in the lateral meniscus, posterior cruciate ligament and collateral ligaments depended on the loads and the flexion angles. This study may help to explain the higher subsequent injury rate of ACL-deficient knee, and to predict the injury risk positions in the joint. Furthermore, this model could shed some light on the mechanism of ACL injuries and the treatments.
    Numerical study on the effect of the floating mass transducer on middle ear sound transmission
    Houguang Liu Na Ta Zhushi Rao
    2010, 42(1):  109-114.  DOI: 10.6052/0459-1879-2010-1-2008-519
    Asbtract ( 1399 )   PDF (1433KB) ( 437 )  
    Related Articles | Metrics
    To overcome some inherent problems in conventional hearing aids, such as low gain at high frequencies, discomfort, and acoustic feedback, floating mass type for middle ear implant system are currently under development. The floating mass transducer (FMT) is the most important component of these devices. For the purpose of investigating the effect of the FMT to middle ear sound transmission, a middle ear finite element model with FMT clamped was constructed. This model was built based on a complete set of computerized tomography section images of a healthy volunteer's left ear by reverse engineering technology. The validity of this model was confirmed by comparing the motion of the tympanic membrane and stapes footplate obtained by this model with published experimental measurements on human temporal bones. The result shows that FMT produces mass loading effect prominently at high frequencies, the force required to drive the incus to the equivalent of 100\,dB SPL is about 8.9 dynes, and placing the clamp point of the FMT close to the incudostapedial joint enhances the driving effect.
    Experimental and theoretical analysis of intelligent control for structural nonlinear vibration
    Luyu Li Gangbing Song Jinping Ou
    2010, 42(1):  115-121.  DOI: 10.6052/0459-1879-2010-1-2008-355
    Asbtract ( 1459 )   PDF (1636KB) ( 557 )  
    Related Articles | Metrics
    Experimental verification of structural dynamics and control strategies is essential for theoretical and numerical study, but only the linearity of the structures was studied in the existing model tests. Therefore, recently there has been an increasing interest in research of establishing experimental model for structural nonlinear vibration. In this paper, MagnetoRheological (MR) rotary brake is used to mimic the plastic hinge of structure so as to analyze structural nonlinear vibration. By means of adjusting the input voltage to MR rotary brakes, different nonlinear behaviors can be detected. Moreover, the dynamical neural network and adaptive fuzzy sliding mode control have been experimentally verified under different nonlinear behaviors by incorporating MR damper into this experimental model. Experimental results show that the structural nonlinear vibration model which is established in this paper can be recovered to initial state without any cost after nonlinear vibration tests. In addition, different nonlinear behaviors can be achieved by changing the input voltage to MR rotary brakes. The intelligent control algorithms are experimentally verified to be suitable for control of structural nonlinear vibration.
    Study news
    xistence condition of one-dimensional self-similar motion of ideal gas
    Xiaoming Lai Baomin Bian Ling Yang Anzhi He
    2010, 42(1):  122-126.  DOI: 10.6052/0459-1879-2010-1-2008-576
    Asbtract ( 1199 )   PDF (713KB) ( 635 )  
    Related Articles | Metrics
    The extension criterion of ideal gas self-similarity motion is not very complete. Hydrodynamic equations are made dimensionless, and then we obtain the basic characteristic quantities solutions of ideal gas one-dimensional self-similarity differential equations described by relative coordinate \xi and distance r. All of them have the same form Y ( \xi ,r ) = y ( \xi ) r^{C_Y }, which means that characteristic quantities for every certain \xi are spatial scale-invariant according to r. It is proved that the spatial scale-invariant is the existence condition of one-dimensional ideal gas self-similarity motion.
    A new method for location leak in long-distance pipeline
    Junhua Li Wenbai Liu Li Cui Zhaochen Sun
    2010, 42(1):  127-131.  DOI: 10.6052/0459-1879-2010-1-2008-222
    Asbtract ( 1336 )   PDF (845KB) ( 479 )  
    Related Articles | Metrics
    For the long-distance pipeline based mainly on frictional head loss, it is necessary to locate the leak point after the leak appears. The researchers have developed many methods for position leak during the past of a few years, such as pressure gradient method, inverse transient method, etc. These methods have their respective advantages and disadvantages for long-distance pipeline. The pressure gradient method is suitable to the pipeline of hydraulic grade line varied linearly. The leak location in the inverse transient method is firstly assumed. Therefore, a new method for location of the leakage points has been developed with the combination of the pressure gradient method and the inverse transient method in this paper. In this scheme, the coefficients of the friction resistance before and after the leakage points, and the coefficients of the leakage flow, are taken as the optimizing control variables, and the difference of water head between the calculated values by waterhammer equations and the measured values at measured sites are utilized as the objective function. The coefficients of the friction resistance before and after the leakage points are optimized through the genetic algorithm, based on which the iterative method is used to locate the leakage points. To verify the new method, the experiment in a water pipe with a length of 48.775m and a diameter of 53mm was set up in the hydraulics laboratory of Dalian University of Technology. The pipeline is made of iron pipe, and is connected to the pump at the upstream end and the water flume at the downstream end. The error value of mean location in this method is about 4%, compared with the experimental data. The experimental results show that the method is quite feasible and efficient to locate the leak in long-distance pipeline.
    Tensile fatigue damage and its mechanism of stitched composite laminates
    Xiaoquan Cheng Kun Yang Renwei Hu Jian Zou
    2010, 42(1):  132-137.  DOI: 10.6052/0459-1879-2010-1-2009-028
    Asbtract ( 1763 )   PDF (1395KB) ( 481 )  
    Related Articles | Metrics
    Stitched laminate is a new kind of composite structure to improve the interface strength of composite. But the in-plane properties of the laminate may reduce because there are local fiber bending in the laminate plane and in the thickness direction, and the damage induced by the stitching threads. A lot of researches have been carried out on the properties of laminates to push this material into engineering applications, such as in-plane tension and compression, compression after impact, hygrothermal properties, etc. But few studies have been performed on stitched composite fatigue properties. The aim of this paper is to comprehend the fatigue damage development and mechanism of stitched laminates under tensile loads. The material of the laminates was uni-weave T300/QY9512 stitched in 0^\circ direction. Un-stitched laminates were studied for comparison with the stitched laminates. Experimental study was carried out on the fatigue performance of stitched and non-stitched composite laminate. The finite element models were established to obtain the stress or strain distributions in the laminates. The mechanism of tensile fatigue damage origination and propagation of stitched laminates was analyzed. The results show that the tensile fatigue damage origination and propagation mechanisms of laminates have been changed by stitching. The stress concentration of in-plane stress \sigma_{x} and interlaminar shear stress near the stitching node plays an important role in damage origination and propagation. Stress concentration of interlaminar stress at the free sides has effect on fatigue performance of stitched laminates. The stress concentration of interlaminar stress at the free sides is the main reason that causes fatigue damage and its propagation of non-stitched laminates.
    A study on the two second-order strain gradient theories
    Jie Zhao Wanji Chen Bin Ji
    2010, 42(1):  138-145.  DOI: 10.6052/0459-1879-2010-1-2008-517
    Asbtract ( 5330 )   PDF (1307KB) ( 574 )  
    Related Articles | Metrics
    There are various strain gradient theories derived from phenomenal theory and nonlocal continuum mechanics. Meanwhile, there are two popular second-order strain gradient theories in which strain gradient term was introduced into the constitutive law with positive and negative signs respectively. In this paper, these two theories are discussed in the aspect of formula derivation, energy expression, and application fields. By using the 18-DOF triangle strain gradient plan element (RCT9+RT9) and axisymmetric element (BCIZ+ART9) which can pass C^{0\hbox{-}1} patch test, the elastic pull-out processes of the reinforced concrete bond specimen and the deformation of a cantilever beam are simulated to compare the limitation of the two theories in the analysis of microstructures.