Modares Mechanical Engineering
Year:2016 | Volume:16 | Issue:8|Papers Count:40

Wind turbines are subject to various unsteady aerodynamic effects. This includes the wind gust and the change of wind direction. In this work، the aeroelastic behavior of a reference horizontal axis wind turbine has been investigated under different wind gusts and yaw conditions. Unsteady blade element momentum (UBEM) theory and Euler-Bernoulli beam assumption were used to rotor power estimations. To take into account the time delay in aerodynamic loads due to a sudden change in inflow conditions، a dynamic wake model was implemented. The ONERA dynamic stall model was coupled into the BEM theory to improve the aerodynamic loads prediction in the unsteady inflow and yaw conditions. To verify this method، the results in the case of steady-state are compared with the NREL 5 MW wind turbine and in the unsteady case are compared with the Tjaereborg test turbine. The results indicate that sudden change in wind speed causes sharp fluctuations in terms of elastic torsion of the blade and other parameters such as rotor power. Increasing in wind gradient can leads to increasing time delay to a new equilibrium. The increase in yaw angle can be contributed to the rotor power and the periodic loads reduction. The method presented here may facilitate improvements in the controller design for wind turbines.

Full feedback data is mostly essential in control design. Measurement of the variation of flexible jointrobot (FJR) actuators is not as easy as the measurement of the changes of FJR links’ angles. Themeasurement of the states is also affected by noise, and the disturbance in the workspace of the robot isnot ignorable. Hence a state observer or a nonlinear estimator is necessary to improve the performanceof the dynamical system. The state-dependent Riccati equation (SDRE) is one of the most promisingnonlinear optimal control methods and estimators. Systematic procedure, simple structure, andincorporating a wide range of systems (under observability condition) are some advantages of SDREmethod. The majority of nonlinear techniques linearize the model, but the SDRE directly uses thenonlinear state space; it is one of the reasons for its precision and flexibility in design with respect toother methods. The goal of this work is to merge the SDRE controller and estimator simultaneously toreduce the state error of the system in presence of external disturbance and measurement noise. So, first, the controller and the observer formulation haves been stated. Then, the procedure has been applied todesign and simulate a 3 DOF robot arm with flexible joints. Next, the process has been testedexperimentally using Scout robot and the simulation results have been verified. Finally, the proposedmethod of this paper has been compared with the optimal sliding mode controller. The results showedthat the behavior of the system is more similar to the real behavior of the robot.

Shape Memory Alloy (SMA) wires are currently employed in robotics as actuators of prosthetic limbsand medical equipment due to advantages such as reducing the size in the application, high power-toweight ratio and elimination of complex transmission systems. In this paper, a fuzzy control system hasbeen designed and implemented for an artificial finger using the SMA actuators. This robotic finger hasbeen designed and modeled with three revolute joints and three SMA wires as the tendon in order foradduction of each phalange of the finger and torsional springs were used to restore them to their originalpositions. The dynamic model of the finger has been simulated in MATLAB/Simulation. Based on thesimulation results, optimal choices of parameters and system features have been obtained and aprototype of finger has been built and tested. Gains of the controllers are set so that the current appliedto SMA wires has minimum overshoot and the output of the system has minimal time to achievestability. The comparison between the simulation results and the actual measured data show that thesimulated model is accurate.

In this paper, an algorithm to detect obstacles surrounding an autonomous vehicle and the method usedto navigate this vehicle on the road were studied. For this purpose, road was divided into cells in lateraland longitudinal directions. The assumption was that some special tools specify the cells positions andthen full and empty-cell corresponding matrix was generated. In this matrix, full cells were displayedwith digit 1 and empty cells are displayed with digit 0. In the next step, by analyzing the matrix inMATLAB®, the vehicle was navigated. In this analysis, first, the position of the vehicle and theobstacles were identified. Then, based on the road conditions and the obstacles positions, requiredorders to move the vehicle were determined. If a lane change is needed, according to the road’scurvature and the distance between the vehicle and the obstacle, appropriate path for the vehicle will bechosen. In this paper, for the first time in autonomous vehicle navigations, the road was considered as a1 and 0 matrix. In this method, the road matrix was updated over time and provides the possibility ofanalyzing the vehicle’s movement. In addition, the algorithm used to solve the problem is very simple.

Piezoelectric actuators (PA) are widely used in electromechanical systems due to interesting propertiessuch as: high resolution, fast response, wide bandwidth, mechanical simplicity, high stiffness. Despitethese unique desirable properties, they suffer from nonlinear behaviors which adversely affect thepositioning accuracy. Among them, hysteresis between applied voltage and actuator position is the mostimportant nonlinearity which can lead to significant error if not compensated. In this study, a slidingmode controller associated with an unknown input observer, which uses the position feedback providedby a selfsensing circuit, is suggested for use in micro positioning applications. The selfsensingtechnique is based on the linear relation between position and charge, which is measured by an activecharge measurement circuit. The advantages of proposed scheme could be summarized as follows. It isa sensorless method which does not need an external position sensor. It does not need any operators tomodel hysteresis or its inverse. It has improved performance in comparison to traditional controllers likeproportional integral (PI) controller. Obtained experimental results demonstrate the effectiveness ofproposed method to use in micro-positioning applications.

In this study, the viscous fingering instability in miscible displacement of Newtonian fluid byViscoelastic fluid is investigated. The Criminale–Eriksen–Filbey (CEF) model has been used as theconstitutive equation. Simplicity and dependence of rheological functions to shear rate are theadvantages of this model. Also, the Carreau-Yasuda model was used to show this dependency. Innonlinear simulation, using spectral method based on Hartly transforms, the effect of rheologicalfunctions of displacing viscoelastic fluid on this instability has been studied. The results includeconcentration contours, transversely averaged concentration profiles, mixing length and sweepefficiency. The results show that, by changing the parameters in order to increase viscosity of displacingviscoelastic fluid, flow becomes more stable. In other words, sweep efficiency is increased and mixinglength is decreased. Also, at first, the sweep efficiency increases with changing the parameters so thatthe first normal stress difference in this type of fluid can be increased and then is decreased withevolution of fingering. However, this factor will have little effect on mixing length. In addition, as wellas viscous fingering, several nonlinear finger interactions such as spreading, coalescence and tipsplitting were observed in simulation of viscoelastic fingering instability.

Free-form surfaces are widely used in engineering applications. These surfaces are complex and withoutrotational symmetry; and for this reason they are inspected using the coordinate measuring machinesequipped with contact sensor which require a suitable sampling strategy. Sampling algorithms are oneof the most important factors of error creation in the accuracy of substitute geometry. In coordinatemeasuring machines, the sampling strategy involves the estimation of the number of sample points (sample size) and identification of their positions (how they are distributed) on the surface. Thus samplepoints should be distributed on the surface using sampling strategies that are appropriate for the surface.Often it is difficult to establish such pieces of information (number and the way of distributing thepoints on the surface) owing to the complex nature of free-form surfaces. In the present work, for thefirst time a new adaptive sampling strategy by particle swarm optimization algorithm (PSO) forsampling from free-form surface is proposed. The proposed strategy was compared with twoconventional strategies and the deviation between substitute geometry and CAD model is extracted. Thesimulation results showed that in the proposed method the deviation between substitute geometry andCAD model is less than conventional methods by 2 to 3 times (depending on the number of points).Therefore, the high efficiency of the proposed method over other methods is concluded.

The aim of this paper is investigation of progressive damage in a metal matrix composite lamina usingcoupling of micromechanical method and continuum damage mechanics viewpoint. Themicromechanical method is a representative volume element- based method known simplified unit cellmethod which possesses the capability of investigating progressive damage and plastic behavior in therepresentative volume elements. The studied damage is isotropic and anisotropic based on continuumdamage mechanics viewpoint. Composite system under investigation is Carbon/Aluminum composite.The matrix behavior is considered as isotropic and elastoplastic and the fiber behavior is transverselyisotropic and elastic. The fiber arrangement within the matrix is regular. The matrix elastoplasticbehavior model is included as bi-linear behavior and solution method is successive approximationmethod. According to available previous studies, Silicon-carbide/Titanium composite system is notedfor validation and comparison with experimental data. Also, the effect of fiber volume fraction on thedamage progression routine is studied. The results show that by increasing the longitudinal andtransverse loadings, the damage variable grows the fiber direction and perpendicular to the fiberdirection and the axial and transverse Young's modulus decrease subsequently. Also, the results provethat in longitudinal loading, considering anisotropic damage, damage progression in the fiber directionis more than its growth, perpendicular to the fiber direction. Whereas, under transverse loading, damagegrowth in perpendicular to the fiber direction is faster.

The use of phase change material to enhance the capacity of energy storage/ release is the subject ofmany new researches on the management of energy supply. Study of these systems is directly related tothe solid-liquid phase-change problem, in which the evaluation of temperature distribution, position ofphase-change front and liquid or solid fraction becomes a basic problem. Study of freezing and meltingprocess with regard to natural convection in the liquid phase is the main purpose of the present paper.For this purpose, a rectangular finned container of phase change material is intended. Fins are used toenhance the heat transfer rate. This fact necessitates the use of immersed boundary condition on thesolid phase. Hence, the melting process considering both the effects of natural convection andmovement of solid phase is studied. The freezing process is also studied taking into account the naturalconvection with no need to impose the immersed boundary condition. Lattice Boltzmann method isused as a numerical method and results are reported based on the dimensionless parameters. Based onthe results, the effects of natural convection are negligible during freezing process, while imposing theeffects of natural convection provides a significant change in the required time for complete melting ofthe phase change material.

A solar-powered robot is a mobile robot powered completely or significantly by direct solar energy. Thesun's energy is converted into electric energy by solar panels mounted on the robot. These solar panelsare required to be light because of the important demands for low-energy consumption. As a result ofthe flexibility of elements of the panels, undesirable low-frequency vibration may occur when the robotmoves on a rough terrain. In this paper, a new method for stabilization of solar panels vibration basedon trajectory planning for articulated mobile robot is presented. The dynamics of solar panels attachedto the robot is derived using Kane’s method. The attitude and configuration of a rover as a function ofthe terrain on which it moves is determined using inverse kinematics of the robot. The attitude andconfiguration of a rover is required to approximate the domain of vibration by derived dynamicsequations. Based on this approximation, a trajectory planning algorithm is presented that can reducevibration with no significant decrease in the velocity of the robot. The proposed method is simulated fora six-wheeled mobile robot with rocker-bogie structure. The obtained results show that the algorithmstabilizes the domain of vibration in allowable area and does not decrease the velocity of the robotsignificantly.

Assessment of strain accumulation due to nonlinear events like creep, plasticity or ratchetingphenomenon has gained importance, as it causes an increase in creep and fatigue damage of materials.Some factors like the magnitude of loading, constitutive equations or the elastic regions around thenonlinear events have an effect on the rate of strain accumulation. The elastic follow-up can explain themechanism of strain accumulation. This phenomenon may occur when a mechanical structure withelastic manner is connected to non-linear events and they are subjected to a displacement load. In thesecases, the high rigidity portion of elastic region of mechanical structure may enhance the force to theregions with low rigidity. So in the local non-linear portion, the strain is accumulated. This phenomenonis proposed as an important instruction in mechanical assessment codes. In this study, the effects ofElastic Follow-up phenomenon on strain accumulation due to elastic-plastic and local creep areinvestigated. So the Elastic Follow-up parameter is defined by the methods which are described in hightemperature assessment procedures (R5). The results revealed that the strain accumulation depends onthe elastic region in structures which is described by the Elastic Follow-up phenomenon.

In the present study, the mixing of two incompressible miscible fluids with different density andviscosity has been investigated in a two-dimensional microchannel equipped with an oscillating stirrerin different excitation frequencies. Although most studies in the field of fluid mixing have studied themixer performance when the two fluids were absolutely identical, the mixing makes sense when twofluids have non-uniformity such as different temperature, concentration or properties. The aim of thisstudy is to evaluate the effect of various properties of the fluids in mixer performance and mixing value.Simulation has been performed in Re=100 and Sc=10, between 0.1 and 1 Strouhal number by usingelement based finite volume method by means of commercial code CFX. Mixer performance has beenevaluated in three different modes: mixing of two identical fluids, mixing of two fluids with differentdensity and mixing of two fluids with different viscosity. The results show that, mixing of the fluidswith different properties leads to change in mixer performance, and has unique performance in eachcase. In comparison with similar properties fluids, mixing of fluids with different viscosity and densityshow lesser tendency in mixing. It has been shown that variation of Strouhal number has lesser effect onmixing index changes. The ratio of maximum mixing index changes to base mixing index in the case ofdifferent density and viscosity is 54.01 and 51.15 percent, respectively, while the value is 577.94percent for the mixing of similar fluids.

In this paper, dynamic behavior of a nano particle on a rough surface in pushing based on the atomicforce microscopy (AFM) was modeled and simulated by using the multipoint contact model. First, amultipoint contact model was extracted for two different roughness profiles of rough surfaces includingthe hexagonal and tetrahedral by combination of the Rumpf singular point contact model with JKR andSchwarz contact models, and the equations of the real contact area and adhesion force were proposedfor multipoint contact of rough surfaces. Then, the dynamic behavior of particles in pushing on therough substrate was modeled by using the new multipoint contact model. Additionally, simulation of theparticles dynamics with radii of 50, 400 and 500 nm in moving on the different rough substrates wasperformed and analyzed, by assuming multipoint, singular point contacts, and flat surface contacts.Results showed that the multipoint contact model, especially in small radiuses of roughness has anessential impact on determination the critical force. Moreover, assumptions of the flatness or thesingular point contact leads to a considerable error in estimating the critical force. Results showedprofiles of rough surface and roughness distribution are very important factors in determining thenumbers of the contact points, and changing the estimated amount of the critical force. In general, theobtained critical force based on the new multi point contact model in comparison with those based onthe flat surface and the singular point contact models, was decreased and increased, respectively.

Abrasive flow machining (AFM) is a relatively new process with low material removal ratio fordeburring, removing recast layers and finishing industrial components with complex shapes among nonconventional machining processes. In this process, the finishing is handled by flowing the compositionof viscoelastic and abrasive particles on workpiece surface, under the pressure of piston. In thisresearch, the abrasive flow machining process of H13 tool steel by applying an external magnetic fieldaround the workpiece to improve the material removal ratio and surface roughness has been investigatedand the effect of magnetic field intensity, abrasive particles mesh and the hardness of workpiece as theinput parameters on the process outputs including surface roughness and material removal ratio wasstudied. Also, the regression model of MRR and surface roughness was developed and variance analysiswas performed. Results of experiments indicated that increase in abrasive-particles mesh leads todecrease in surface roughness and material removal ratio and increase in magnetic field intensity causesmaterial removal ratio to increase and surface roughness to decrease. Also, the material removal ratio isdecreased with increasing workpiece hardness and in the same condition better surface finish wasachieved in the case of harder workpiece.

In this paper, the performance of a single-axis attitude control with pulse-width pulse-frequency (PWPF) modulation is enhanced using a modified proportional-integral-derivative (PID) controller for arigid satellite with on-off thruster actuators. For this purpose, the well-known observer-based PIDapproach is utilized. The on-off thruster actuator is modeled with a constant delay followed by asecond-order binomial transfer function. The modulator update frequency is limited to 40 Hz as an inputto the on-off thruster actuators. In this study, the design criteria of pointing accuracy, overshoot of theattitude response, fuel consumption, and the number of thruster firings are considered for a step externaldisturbance (with different values). The parameters of the observer-based PID controller are tuned using parametric search method. Simulation results show that the fuel consumption and settling time of theobserver-based approach are considerably decreased with respect to those of PID controller with PWPFmodulator. Moreover, the overshoot of the observer-based approach is omitted. Finally, the robustnessof the observer-based modified PID controller is investigated in presence of uncertainties in satellitemoment of inertia and thrust level of on-off actuators.

Differences in the persons’ individual parameters such as age, gender, weight, height and basalmetabolic rate have a significant effect on the human body thermoregulation. Therefore, using thehuman thermal models that were developed on the basis of large human population does not lead toaccurate results for specific individuals. Because the individual parameters have not been considered instandard thermal comfort models and also available individual and local models are so complicated inapplications, nowadays, the necessity of developing a simple and accurate individualized model is vital.In this study, some physiological parameters such as: body fat percentage, subcutaneous fat layerthickness, body heat capacity coefficient and tissue conductive resistances have been modeled fromreadily-available external measurement of individuals and these parameters are incorporated into threenode-model algorithm structure to predict individual variations in thermal response between individuals.Three-node thermal comfort model is based on Gagge’s standard model that has accurately estimatedthermal sensation of the bare and clothed parts of the body. The model has been verified against theanalytical and experimental results where good agreement was found. In conclusion, the results indicatethat the mean error in prediction of skin temperature is decreased from 1.2 oc for three-node model to0.4 oc for the new individual model.

Electrical discharge machining is one of the most common and widely used machining processes formachining hard metals and alloys which has low machinability by traditional machining methods. Dueto the thermoelectric nature of this process, changes in metallurgical and mechanical properties ofmachined surface and development of residual stresses in components are inevitable. In this researchmachining of Ti-6Al-4V titanium alloy is conducted by ultrasonic assisted electrical dischargemachining process and the effects of ultrasonic vibration of tool on the machining efficiency, surfaceintegrity- such as surface micro-cracks, residual stress and surface hardness have been evaluated.Machined surface was imaged by scanning electron microscopy imaging to study the size anddistribution of surface micro-cracks. Residual stresses along the depth of the machined surface, evaluated using Nano indentation technique and hardness of discharged surface are measured using amicro hardness measuring instrument. The results show that applying ultrasonic vibration increaseselectrical discharge machining process efficiency (about 90%), reduces the amount and size of surfacemicro-cracks, changes residual stress distribution and decreases the amount of it (average 17%).Increase of surface hardness caused by ultrasonic assisted electrical discharge machining process is 13%more than the traditional electrical discharge machining process.

In this paper, the effect of different workpiece geometries on the properties of welded Al-7075-T6 partsin rotational friction welding process has been investigated by using experimental and finite elementapproaches. Welding process is continuous drive friction welding. In this research, the samples diameterwas selected equal to 25 mm and the samples length was selected equal to 75 mm. The processvariables such as friction time (t1), forging time (t2), friction pressure (P1) and forging pressure (P2)were assumed to be constant, whereas rotational speeds were variable and selected equal to 2000 rpmand 2500 rpm. Three cylindrical parts with different cross sectional geometries were adopted as threesamples. Finally, to verify the accuracy of numerical analysis, the experimental and thermo-mechanicalsimulation results were compared and good agreement was found between them. Since an experimentaltest is a time consuming and a costly process, it was decided to obtain more results by using analternative method such as finite element simulation technique. Results of this study showed thatchanging in front geometric of workpiece is an effective factor for tensile strength, length of workpieceand generated flash in welded area and by changing this factor, properties of welded area can beimproved.

The nonlinear vibration of sandwich viscoelastic plates under wide-band random excitation is investigated. The main attention is put on the influence of the one-to-one internal resonance، arisen from the close natural frequencies of the asymmetric modes of a near-square plate، on the response. The multi-modal response and the on-off intermittency phenomenon are especially considered. The mathematical modeling of the mid-layer is based on the moderate transverse shear strains and rotations، which have led to both geometrical and material nonlinearities. For the nonlinear constitutive equation of the mid laye a single integral viscoelastic model is used. The displacement field in the thickness direction is also assumed to be linear for the in-plane components and quadratic for the out-of-plane components. Moreover، the Kirchhoff theory with the von-Karman nonlinearities are used for the outer layers. The solution is initiated by applying the perturbation method along with the Galerkin’s method to obtain integro-differential ordinary equations in time. These equations are then، solved using the Gaussian and non-Gaussian closure methods and the results are used to investigate the occurrence of the bifurcation with the aid of the Pseudo-arclength continuation method. Numerical results are presented for the multi-modal response and the minimum excitation intensity required for the nonlinear interaction between asymmetric modes.

The aim of the present work is the investigation of polypropylene/graphene oxide nanocamposites. Inthis work, the reinforcing effects of the graphene oxide nanoparticles on the mechanical and thermalproperties of the nonpolar polypropylene are examined. There are two main challenges to improve theproperties of polypropylene by graphene oxide nanoparticles. First, the nanoparticles do not havesuitable dispersion in polymer matrix. Seconde, there is not sufficient adhesion between nanoparticleand polymer chains. In this study, the graphene oxide (GO) surface is modified by a linear alkyl chainvia a bimolecular nucleophilic substitution reaction between the oxygen groups of GO and reactants topromote the homogeneous dispersion of GO in the organic solvent and increase the interfacial adhesionbetween the graphene oxide and polymer matrix. The presence of the alkyl groups on the surface ofgraphene oxide nanoparticles is characterized by FT-IR. To prevent the AGO aggregation in thepolypropylene, the solution-blending method is used to prepare the nanocomposites with 0.1, 0.3, 0.5wt% AGO. The SEM images confirmed the appropriate dispersion of the graphene oxide in thecomposites. The stress-strain analysis, dynamic-mechanical thermal analysis (DMTA), and thermalgravimetric analysis (TGA) are performed to investigate the mechanical and thermal properties ofnanocomposites. The results demonstrated that the Young’s modulus of the polymer is improved by 20, 30 and 34% by the addition of 0.1, 0.3 and 0.5% AGO, respectively. Also, the 10% mass losstemperature for 0.1, 0.3 and 0.5% AGO nanocomposites compared to neat polypropylene increased by2, 8 and 12 C 0, respectively.

The main purpose of this paper is modeling of the free corner effect of cross-ply and angle-ply graphite/epoxy composite laminates using finite element method based on global-local method. The global area is modeled by first order shear deformation theory and the local area، in the free corner vicinity، is modeled by the Reddy's layer-wise theory. Using this method provides the possibility of analysis of thick angle-ply and cross-ply laminates. The cross-ply and angle-ply laminates are subjected to uniform thermal and extension loading، respectively and the effects of the free edge and free corner interlaminar stresses are investigated. The presented results verification is performed via available results in the previous studies which show good agreement. The present study results show that when the cross-ply laminate is subjected to thermal loading، the interlaminar stresses distribution is uniform in both length and width of the laminate. However، for the uni-axial extension loading، the interlaminar stresses possess different distribution in the two directions of the laminate. Also، results demonstrate that in angle-ply laminates under extension loading، the free corner effect increases by increasing fiber angle and the maximum interlaminar stresses occur in 30 degree plies in the free corner vicinity. Moreover، results prove that the effects of the free edge and the free corner are almost similar in layers with fiber angle less than 30 degrees. Parametric study on the thickness and stacking of the laminate layers display that both parameters have a significant influence on the interlamianar stresses at the free corner.

Satellite Thermal Control subsystem has the responsibility of maintaining the temperature of othersubsystems in an allowable range. The purpose of this paper is to design an optimal thermal controlsubsystem of a satellite. In order to achieve this goal, at first a software was developed and validated forsatellite thermal analysis. Receiving orbital data and Satellite’s properties, the software simulates theposition of the satellite in any desired orbit and calculates the input and output thermal flux. Meanwhile, the temperature of each side of the satellite is calculated in cold and hot cases. Finally, it calculates theminimum and maximum temperatures of the satellite. A combination of three commonly used thermalcontrol methods in small satellites was used. Insulation thickness, thickness of radiator cover, and thepower of heater are considered as design parameters and allowable temperatures of surfaces (minimumand maximum allowable temperatures) are considered as design constraints. A weighted function ofmass, cost, and power consumption of thermal control system are chosen as objective function whichc an be representative of the cost. Sequential Quadratic Programming as a powerful method in nonlinearoptimization was used to optimize thermal control properties. The results demonstrated that theobjective function improved dramatically compared to the initial design. High speed, appropriateprecision, and extensibility of this software in thermal control subsystem design of a vast majority ofsmall satellites, makes this research valuable. Therefore, this software could be integrated as the thermalcontrol subsystem design module with multidisciplinary design optimization of satellites.

In this study the steady-state dynamic of a linear, homogeneous, un-damped string, coupled with alocally connected spring-dashpot system is analytically investigated. Both ends of the string areassumed to be excited with identical and synchronous harmonic motion. It is shown that the damperintroduces mode complexity and leads to frequency shift between the peak amplitudes in differentlocations of the string. Also, it causes phase variations which indicates mode complexity domain. In thisstudy, it is shown that there are different combinations of spring and damper constants in which themode complexity attains its maximum level. Surprisingly, the combination is unique in each givenexcitation frequency ratio. In this situation, the damping constant is bounded in a specified range, butthe spring constant is increased as the excitation frequency ratio is increased. In such case, all vibrationnormal modes of the string are completely destroyed and, in turn, traveling waves are formed. Also, it isshown that the damping constant which leads to the maximum frequency shift is not necessarily equal tothe one that introduces the maximum mode complexity.

One of the remarkable concerns in Shallow arches’ behavior under lateral loading is snap-through, aphenomenon which can make the structure collapse or displace to another stable configuration.Introducing functionally graded materials in recent years has led to some interesting results, forinstance, using functionally graded materials in shallow arches can produce structures with favorablestability properties. In this work, we investigate dynamic stability of the pined-pined functionallygraded sinusoidal shallow arch under impulsive loading. Material properties vary through the thicknessby power law function. Nonlinear governing equations are derived using Euler-Bernoulli beamassumption and equations of motion are expressed by a nonlinear differential-integral equation. Thesolution utilizes a Fourier form of response. The procedure to analyze dynamic stability followed hereuses total energy of the system and Lyapunov function in the phase space. We find the stable regionagainst dynamical snap-through under material properties’ variation in the thickness direction ofshallow arch. We also proceed to find the sufficient critical load in order to make the dynamical snapthrough occur. The results are analyzed in detail and illustrated in some diagrams.

micro-mechanical model based on the representative volume element (RVE) is presented to study thetime-dependent and creep behavior of fibrous composite material. To this aim a finite element model ispresented for analysis of creep behavior of material in multi-axial creep are presented. The generalizedplane strain condition is employed to model the behavior of the RVE in axial and transverse normalloading. The governing equations of the problem in the RVE are discretized using the presented finiteelement method and the stiffness and force matrixes are presented. Appropriate boundary conditions areimplied to the RVE in order to consider the transverse and axial loading conditions including creepbehavior. The Euler explicit method is employed to solve the discretized equations in the time domain.The distribution of micro-stresses and the effect of creep in re-distribution of the stresses are studied.The steady state creep behavior of composite in macro-mechanical scale is investigated by analysis of the micromechanical behavior of the RVE. The macro-mechanical creep behavior of metal matrixcomposite in axial and transverse loading is predicted from the presented micromechanical model.

Unbalance mass and imperfect bearings are the main sources of vibration in rotor dynamics systems.One way to decrease and control a rotor vibration is the use of magnetic absorbers. The magneticabsorber is used to control the position of the rotor and reduce its vibration. In this study, by applyingthe dynamic absorber system force and creating two new natural frequencies, the magnetic absorberbrings the system out of the resonance. Moreover, in order to decrease the vibration amplitude, twodifferent types of dynamics absorbers are designed in which they are checked by the magnetic absorberin a specific range of rotational frequency. In magnetic absorber controller system, the continuous forcewhich is applied to the main system by mass absorber is restored in sixteen levels discontinuously. It isseen that the vibration amplitude is reduced 13% in the area of natural frequency in comparison to themagnetic absorber with discontinuous force. In this paper, two different mass ratios are considered foreach one of the two absorber systems. It is observed that in the case of dynamic absorbers with highermass ratio, rotor vibration amplitude and the maximum force amplitude of the dynamic absorber systemdecrease. This issue can increase the accuracy of magnetic absorber system in the renewal of thedynamic absorber system force and reduce consumed electrical energy of the control system as well.

A RTICLE I NFORMATION A BSTRACTOriginal Research PaperReceived 19 June 2016Accepted 12 July 2016Available Online 28 August 2016Effects of different inlet/outlet arrangements on thermal performance of porous microchannel heat sinkMCHS of any geometry has not been studied yet. In this investigation, the effects of utilization of fourdifferent inlet/outlet arrangements on electronic chip cooling utilizing trapezoidal MCHS with porousmicrochannels with porosity of 0.88 have been studied numerically. For this purpose, three dimensionalsimulations of laminar forced convection flow in microchannels and conduction in solid parts of MCHSby applying constant heat flux of 150 kWm -2 at its base plate have been performed utilizing the finitevolume method and the commercial Ansys-CFX code. The results show that the A- and B-typearrangements, for which the inlet and outlet are in direction of flow in the microchannels, have a betterheat transfer performance, smaller thermal resistance and provide more uniform temperaturedistribution in the MCHS base plate. The results indicate that using porous media is effective inreducing the MCHS base plate temperature and in this regard the D-type arrangement has the bestperformance among the heat sinks studied. Considering both the positive effect of using porous mediaon increasing the heat transfer coefficient and its negative effect on increasing the required pumpingpower, the A-type arrangement has the best performance.

The large amount of diesel engine waste heats compels researchers to design systems that utilize theengine waste heat to provide the cooling demand of the heavy-duty vehicles and improve the engineefficiency. Considerable advantages of adsorption cooling system lead to them being selected for thispurpose. Coolant and exhaust gases are the main sources of waste heats of diesel engines and using eachof them to drive the adsorption cooling system requires its own equipment and working pair. In thispaper, a detailed numerical model has been developed to examine the performance of the adsorptioncooling system driven by the coolant and exhaust waste heats with the working pairs of silica gel-waterand zeolite13x-water, respectively. An identical absorbent bed and ambient conditions have beenemployed to compare the performance of both systems to identify the more appropriate system. Theresults show that exhaust driven adsorption cooling system has a better capability to meet the vehiclecooling demand. Moreover, the performance of both adsorption cooling systems was examined undervariable ambient condition. Results indicate that increase in ambient temperature leads to an almostlinear performance drop in both systems that is more considerable in the coolant-driven adsorptionsystem.

An undesirable factor that affects the dimensional precision and final shape of metallic parts producedin cold forming processes is springback phenomenon. An analytical model is introduced to predictspringback in U-shaped bending process of DP780 dual phase steel sheet. This analytical model is basedon the Hill48 yield criterion and plane strain condition. In this model, the effect of forming history, sheet thinning and the motion of the neutral surface on the springback of U-shaped bending process istaken into account. The anisotropic nonlinear kinematic hardening model is used to consider the impactof complex deformation, including stretching, bending and reverse bending. This model is able toinvestigate the complex hardening behavior of material such as Bauschinger effect, transient behaviorand permanent softening. The effect of the sheet holder force, the coefficient of friction, thickness, material anisotropy and hardening parameters on the sheet springback is studied. It can be seen thatanalytical model which is presented in this paper has good accuracy in the springback prediction incomparison with FEM method and results are close to experimental data. The results show that the sheetholder force, the coefficient of friction, thickness and material anisotropy have considerable influenceon the springback prediction. Since, the sheet experiences a reverse loading during the forming process, hardening parameters of the material have a significant influence on the springback prediction. It can beseen that the Bauschinger effect has more influence on the springback prediction than the permanentsoftening and transient behavior.

In this paper the dynamic behavior of a rotating system which includes rotor (shaft), ball bearing anddisk in stationary condition and different speeds is investigated. There are nonlinear characteristics inthese systems which make the linear modeling inaccurate. So, in this paper the nonlinear dynamicequations of the system are derived and solved. To derive the equations of the system, Hamiltonianmethod is used, and complex coordinate transform is employed to reduce the number of equations.After solving the equations, to investigate the vibrational properties of the system, time responsediagram, dynamic orbit, frequency response, and mode shape of the rotor is plotted. To validate theanalytical results, finite element method by ANSYS (workbench) software is used. There is goodconformity between the analytical results and finite element results in resonance frequencies of thesystem in the first three modes which indicates the sufficient accuracy in nonlinear modeling. It can beconcluded from nonlinear modeling that the decay rate is negative for the all modes, which indicates thestability of them. Also, the maximum vibration amplitude in the bearing and rotor occurs in third andsecond modes respectively. Unbalance phase difference of 90 degrees in two discs causes the excitationof all three frequency modes, whereas by unbalance phase difference of 0 or 180 degrees in two discs, only the odd modes (first and third) and the even modes (second) are excited respectively

Many models have been applied to slug flow using laminar flow condition. The results obtained from these models are notconsistent with the physical behavior of slug flow. Furthermore, discussion on the turbulent models is very rare or not related tosuch flow regime. The slug regime is a complicated regime with shear flow and high strain in addition to some vorticity at somesections of the flow.In the present attempt, in the first stage, the turbulent models differences, the initial assumptions todrive, privilages and shortcomings have been considered with details. Then, its consistency with thephysics of slug flow was analysed with high accuracy. In the second stage, simulations using differentturbulent models were conducted. The obtained results were compared to each other and with theexperimental results of other investigators. Finally, the most consistent model with the physics of theslug flow was selected. The turbulent model of RNG k-ε showed more reliabilitycompared to otherturbulent models. Thus, it was selected and used to obtain slug flow behavior with higher accuracy. Theparameters as pressure distribution during slugging, slug mixture velocity, slug initiation time andposition from the duct inlet with RNG model were conducted and presented with detailed explanations.

In this paper, to improve the accuracy of the one-mass and three-mass inverted pendulum models, which have been used for generating real-time walking patterns for biped robots, we propose a novelmodel based on the three-mass inverted pendulum. The proposed model employs an approximation ofmoment of inertia of the swing leg to improve the accuracy of the three-mass inverted pendulum inestimating dynamic behavior of the robot. In order to show the significance of the proposed model, trajectories for the Center of Mass (CoM) are obtained using the three models, based on a desired ZMPtrajectory. The task space trajectories are then mapped into the joint space, using inverse kinematics.Having the joint space variables, the actual ZMPs for the three obtained walking patterns are computedand compared. This comparison shows the advantages of the proposed model in estimating dynamicbehavior of the robot well, especially for walking with relatively high speeds. The kinematic anddynamic properties of the models in this paper are based on the humanoid robot SURENA III, whichhas been designed and fabricated in the Center of Advanced Systems and Technologies (CAST), University of Tehran.

Mechanical behavior of articular cartilage is affected by many factors. Inhomogeneous distribution ofproteoglycans and collagen fibers through the thickness causes some depth-wise behavior. Mechanicalproperties directly affect stress and deformation of the tissue. In previous studies complexities andvariation in mechanical properties were ignored. The aim of the present study is to create a model closeto real anatomy of articular cartilage in knee joint and to simulate its behavior under dynamic gate in thestance phase.A 3D finite element (FE) model was created. It was constructed considering femur and tibialcartilages as well as medial and lateral meniscus. In the FE model, a nonlinear isotropic viscoelasticmaterial model was used for cartilages and linear anisotropic elastic one was chosen for meniscuses.Morever, cartilages were assumed saturated. Numerical simulations on the model showed that peak ofmaximum principal stress occurred in superficial layer. It was decreased through thickness. Thisexplains the existence of osteoarthritis in the exterior layers. The present study showed that hydraulicpermeability variation in cartilage as a strain-dependent variable was negligible in dynamic loading.Also, results showed good agreement with experimental ones.

The interaction between vortical structures and spherical particles or droplets is of practical issue intwo-phase flows. The interactions bring major changes in the flow field particularly when coupled withparticle rotation. It is observed that the heat transfer rate is significantly influenced during the time thatthe vortices’ cores are in the vicinity of the particle. In this paper, transient heat transfer of a rotatingspherical particle interacting with a pair of vortices in incompressible and viscous flow is studied usingnumerical solution of the Navier-Stokes and energy equations in the range of 20≤Re≤100 and nondimensional rotational velocities 0≤ Ω ≤1; by computational code which has been developed by theauthors. In order to ensure the accuracy of the calculation, the results are compared with numerical datareported in the literature and good agreement between results was observed. Then the effect ofcirculation direction of two vortices interacting with a particle by spin on its heat transfer rate wasinvestigated. Also, distribution of heat transfer coefficient at the particle surface with separate rotationaround three different axes in two cases of interacting and non-interacting with vortices is given and theresults of heat transfer coefficient are presented. The results show that particle rotation for Ω ≤0.5, inboth presence and absence of vortices in flow field has negligible effects on the particle heat transferrate; however, with increasing particle spin significant effects on heat transfer coefficient has beenobserved that, due to the circulation direction of vortices, different amounts are obtained.

In this paper, motion of a flexible membrane and hydrodynamic interaction of multiple membranes in amicrochannel are simulated by developing a computer code written in C. The membranes areconsidered as flexible boundaries immersed in the fluid. First, a single biconcave shaped membranewith high rigidity is considered. Due to the rigidity of the membrane, tumbling motion occurs andvertical displacement becomes oscillatory. Then, the effects of initial position of a circular membraneon its deformation, vertical velocity and displacement are investigated. It was observed that as the initiallocation of the membrane approaches the channel’s central axis, its vertical displacement and velocitydecreased, but its horizontal velocity component increased. Finally, the simultaneous motion of multiplemembranes in a microchannel and their interaction with each other and with flow are evaluated. Themembranes do not collide and hence the collision mechanism is not modeled. It was found that theupstream membrane experienced greatest deformation and the greatest force was exerted on it by thefluid on it. In addition, simultaneous presence of multiple membranes would result in a reduction in theflow velocity. The current numerical results have good agreement with the available valid numericalresults.

In this study, the effects of geometrical parameters of 6-DOF Hexa parallel robot on kinematic, anddynamic performance indices are investigated and its structure is optimized using the intelligent multiobjective Bees Algorithm. In this way, after describing the structure and specifying the geometricalparameters of the robot, inverse kinematic relations of the robot are obtained. Jacobian matrix that mapsvelocity from joint space to Cartesian space is developed. Mass matrix is obtained from calculating thetotal kinetic energy of the manipulator in terms of the actuated joints vector. Inverse of the homogenjacobian-based condition number is considered as an index to evaluate the kinematic dexterity. Basedon mass matrix as relation between acceleration vector of the end effecter and torque vector of actuatedjoints, dynamic dexterity index is presented. Using the multi-objective Bees Algorithm and consideringdynamic and kinematic performance indices in a pre-determined workspace as the objective functions, structure of Hexa parallel robot is optimized. In this way, the proper geometrical constraints such aslimitation of universal and spherical joints, and singularity avoidance constraints are considered. Paretofront of the multi objective optimization of the robot is drawn. Diagrams of the kinematic and dynamicperformance indices variation in the workspace and the effects of geometrical parameters variation onthem are presented.

This paper deals with atomistic modeling of nanomechanical behavior of actin monomer. The majorcytoskeletal protein of most cells is actin, which is responsible for the mechanical properties of the cells.Actin also plays critical mechanical roles in many cellular processes which give structural support tocells and links the interior of the cell with its surroundings. Based on the accuracy of atomistic-basedmethods such as molecular dynamics simulations, in this paper, we perform a series of steeredmolecular dynamics simulations on both ATP and ADP single actin monomers to determine theirintrinsic molecular strength. The effect of virtual spring of steered molecular dynamics on themechanical behavior of actin monomer is investigated. The results reveal increasing the virtual springconstant leads to convergence of the stiffness. The stiffness of ADP actin and ATP actin is calculated as215.16 and 228.24 pN/Å, respectively. The results also show higher stiffness and Young’s modulus forATP G-actin in comparison to ADP G-actin. In order to compare the behavior of ATP and ADP Gactins, the number of hydrogen bonds and nonbonded energies between the nucleotide and the proteinare analyzed. The obtained persistent length is 15.61 mm which is in good agreement with the otherreported literature values.

In the present study, the incompressible flow through highly heterogeneous porous media is modeled bythe Multi-resolution Multi-scale Finite Volume (MrMsFV) method. In order to focus on the effects ofthe absolute permeability structure on the accuracy and performance of the MrMsFV method, the singlephase flow is considered and the effects of the gravity and variation of fluid viscosity and density areignored. The accuracy of the MrMsFV method is examined by comparing its numerical results withthose of the standard finite volume method. These permeability fields are extracted from the tenthcomparative study problem of the Society of Petroleum Engineering. For the permeability fields inwhich the permeability varies smoothly, it is shown that the MrMsFV method produces acceptableresults. On the other hand, the numerical results along with mathematical analyses show that theMrMsFV method may produce pressure fields with unphysical peaks for channelized permeabilityfields. In these cases sufficient conditions for the monotonicity and boundedness of the solution areviolated. In fact, the coarse scale transmissibilities may be computed in such a way that the coefficientmatrix of the coarse scale pressure equation does not become a so-called M-matrix.

A hexapod machine tool with a parallel structure has six degrees of freedom. This machine has a highdexterity unlike traditional machine tools. The hexapod can be used in machining free form surfaces.Free form surfaces are widely used in today’s industries. These surfaces are much encountered in auto, aerospace and mold design industries. Consequently, machining of these surfaces has attracted theattention of researchers. In this field much research has been done on five axis machine tools. In thispaper machining free form surfaces with hexapod machine tool has been investigated. The main topic ofthis paper is the feasibility of using hexapod as a machine tool table and machining with it. First, theinterpolation of free form surfaces for parallel structure machines is explained. Then NURBS curvesand surfaces are described and its formulation in matrix form is explained. After that, extractinginformation of free form surfaces with NURBS formulation is explained. Subsequently, someexplanations about preparation of machining are given. Finally, two free form surfaces designed inCatia have been machined with the developed hexapod machine tool.