مجله مکانیک سازه ها و شاره ها
سال پنجم شماره 4 (زمستان 1394)

Tunnel fires are responsible for many fatalities in recent decades and this field of study has received an extensive effort by researchers, investigators, students and so forth. Removal of the generated plume and high temperature is the great interest of investigators. In the current study, a numerical test has been performed on a rectangular cross section tunnel with a pool fire at the middle part using FDS5.5. Obtained results were compared with previous results. Then vertical distributions of Carbone monoxide and temperature stratification are presented for various heat release rate of fires and various tunnel aspect ratios and inclinations which have been reported in dimensionless form using their maximum value beneath the ceiling. Results indicate that the vertical values of CO decay faster than temperature values. These profile also helps us to find out about thickness of the plume where one can be aware of smaller thickness for higher aspect ratio.

In this paper, two types of flows are modeled using the Isogeometric Analysis (IA) method. The first problem is to find the velocity distribution of uniform flow in a sloped channel and the second is irrotational flow around circular and rectangular obstacles. The formulation is derived and its differences with the finite element (FE) method are explained. In the IA method, the unknown function of the governing differential equation and the domain boundaries are approximated by NURBS (Non-uniform Rational B-Splines). Due to the ability of NURBS in constructing curves and surfaces with high precision, channels with complicated boundaries can easily be considered. The IA results are compared with the standard finite element and the accuracy is demonstrated by several examples. Furthermore, the effect of some of the IA parameters such as the irregularity of the control point grid, different knot vectors and the number of control points are also discussed.

In this research, the effect of viscosity of fluid on forming of flat head cylindrical cups by using hydrodynamic deep drawing process assisted radial pressure, was studied experimentally and also by finite element analysis. In this paper Firstly, to determine forming parameters such as maximum pressure and applied pressure path, process was carried out by numerical simulation by finite element method. Then by using several types of fluid with different viscosity as pressure carrying medium, several experiments were carried out. For every fluid, thickness distribution in final part and maximum punch force were investigated. By comparison of experimental results with numerical data, it has been found that results from numerical simulation were in good agreement with the experimental results. In addition in this paper it is shown that increasing the fluid viscosity leads to increasing of maximum punch force. In addition increasing fluid viscosity doesn’t have any effect in maximum blank thickening in the cup wall zone. Finally, in this research it is shown that by selecting fluid with appropriate viscosity, final parts with more uniform thickness distribution could be achieved.

In most simulations conducted by researchers in the turning process, damping of system is considered from modal analysis. Indeed, in these studies the contact of the tool with the workpiece is not taken into account that can be reduced the accuracy of the simulation. In this paper, the process damping of the turning process is defined by using time series analysis. The recommended method is based on comparison with the correlation dimension of the simulation to experiment of displacement signal in the feed direction . Therefore, the variation of process damping based on variations feed and spindle speed is calculated. The results simulation of displacement of tool in feed direction in MATLAB software against displacement of tool in experiments, indicate significant accuracy of the proposed method in determinig displacement the tool in the turning process. The results of suggested approach can be used for predicting surface roughness of the manufactured workpiece by using turning process.

Collocated Discrete Least Squares (CDLS) meshfree method, is used for the numerical simulation of the Sain-Venant torsion problem and obtaining the shear stresses in the sections with irregular and complex boundaries. In this paper a matrix formulation is applied in discretizing the governing equations, which make the procedure easy to code and efficient in calculation. In the applied method, a limited number of neighbor nodes are considered in the producing radial basis shape functions, which make the matrices sparse and prepare a suitable condition for sparse matrices algebra and applying available subroutine in this case. In order to validate the proposed method in the analyzing stresses due to torsion, firstly an elliptic section is considered , which in this case, there is an analytical solution. In the second problem, a thin walled section is considered and solved by the theory of Saint-Venant in torsion and results are compared with the closed solution obtaining from the theory of the torsion of thin walled sections. In the third problem, torsion in a hollow elliptical section is solved. Finally, the torsion stresses in sections with more complex geometry is considered.

In this study, the effects of temperature, relative humidity and air velocity on the effective coefficient diffusion on four different sample clay ceramics were investigated. A full factorial design of experiments on four cubic samples was performed. In each trial mass and volume reduction were measured using an electronic balance and a vision system and the resulting data were recorded and the drying curve was plotted. Effective diffusion coefficient of moisture using drying curves were calculated using of newly developed Firefly algorithm in which the error defined as the difference between the analytically values and the experimental values were minimized. Statistical analysis and analysis of variance showed that in calculation of the effective diffusion coefficient; the velocity, the temperature and the humidity are independent of each other. The greatest impact on the effective diffusion coefficient is related to the velocity, Followed by temperature and environmental humidity respectively. Finally from the drying parameters the regression of the diffusion coefficient was obtained.

In this paper, at first a Shape Memory Alloy (SMA) actuator is modeled by Brinson model. Due to high efficiency of Brinson model in modeling SMA behavior, this model is studied in detail in all regions of temperature-stress plane. Then, an SMA actuator, that is composed of a SMA wire connected to a linear spring, is modeled. Since in many SMA actuators, some parameters, like spring stiffness in this case, cannot precisely determined or it is changed by the environment situations, an Extended Kalman Filter (EKF) is used to estimate these parameters. In this filter, an initial guess and its error covariance must be determined. But since this error covariance can’t be determined precisely, the effect of initial error covariance inaccuracy on the parameter estimation is analyzed. Also, since in actual cases the spring may have nonlinear behavior, a simulation is performed by using a nonlinear spring while the estimation is carried out by using linear spring model. These analyses show that EKF can estimate parameters of an SMA-actuated actuator quickly as well as accurately. Also these estimations are robust against the model and initial guess uncertainty.

This paper presents a novel design for the robust impedance control of a lower-limb rehabilitation robot by using fuzzy parameters and voltage control strategy. Selecting the impedance parameters has been a challenging problem, thus they are given by fuzzy systems in the proposed design. Compared with the previous designs, the novel impedance control approach is not dependent on the mechanical model of robot. Designing the controller is based on the voltage control strategy which differs from the common torque control strategy. Compared with the designs based on the torque control strategy, it is simpler, less computational and more effective. In addition, the controller considers the actuators. The proposed approach is robust adaptive against uncertainties by using a gradient decent algorithm. The stability of control system is proven and the efficiency and superiority of the robust impedance control with fuzzy parameters over the robust impedance control with constant parameters are shown by simulation results.

Fiber Metal Laminates (FMLs) are consist of metal alloys and polymer composites. The benefit characteristics of both metals and composites consist in FMLs. In this experimental research work, four sets of FMLs were fabricated by a hand lay-up method and impact resistance induced by drop-weight instrument on different lay-up configuration was studied. In all sets, the top and bottom layers were Aluminium 2024 - T3 sheets. between the top and bottom Aluminium sheets, there were two composite layers. two sets, they are epoxy resin reinforced with carbon and aramid fibers fabrics. In the other two sets, the composite layers are epoxy resin reinforced with glass and aramid fibers fabrics. Experimental results showed that the best impact resistance is related to the FMLs consist of aluminum sheet/ epoxy resin- aramid fibers fabric / epoxy resin- glass fibers fabric/ aluminum sheet and it has lowest impact damage and lowest permanent central deflection.

In this study, elastic support assumption was used for crack detection in beams. At first, the model of beam was updated by assuming that boundary conditions of beam restrained by rotational and translational springs. Experimental modal data of intact beam are considered as input in this stage. An inverse problem was used to determine unknown values of support springs’. At the second stage, developed model for the beam at the first stage was then employed. Crack detection has been done using an inverse problem. Natural frequencies of cracked beam are considered as input in this stage while severity and location of crack are unknown. Inverse problems are solved by using firefly algorithm where this algorithm show suitable convergence rate. Efficiency of proposed method for single and multiple crack detection in beams using experimental data were compared with other studies. Results indicate that proposed method has better accuracy regarding crack severity and location detection.

In this paper, by combining the merits of model reference adaptive control, sliding mode and fuzzy control a novel control approach is proposed to address the stabilization and tracking problem of Lorenz chaotic system without chattering phenomenon in the face of structure and unstructured uncertainties. In this control scheme an adaptive fuzzy system is used to estimation unknown function. As a result, there is no need to calculate the bound of unknown function and chattering phenomenon is attenuated. One of the advantages of proposed method is that the behavior of close loop system is similar to proposed model reference in the presence of uncertainties and disturbances. As a result, the shortcoming of adaptive control in the presence of unmodeled dynamics and external disturbances is compensated. Another advantage of proposed approach is that chaotic systems have unpredictable behavior and extreme sensitivity to initial conditions. The stability analysis is verified and the effectiveness of the proposed method is compared with backstepping method through simulation. Numerical simulations illustrate that proposed control approach is superior to backstepping method already published in literatures in overcoming uncertainties.

The analysis of sandwich beams with FG face sheets loaded by central indentor in various temperature conditions is carried out in this paper. Property distribution in the FG face sheets is according to power law function of FGMs and all properties of them are temperature dependent. In this model, First order shear deformation theory is used for the FG face sheets while three-dimensional elasticity is used for the flexible core. Two spreading length scales are introduced and calculated for defined sandwich beam, that characterized the behavior of sandwich beam under local loads. These spreading length scales, which are two functions of the beam material and geometrical properties, characterize the length over which a load on the upper surface of a beam is spread out by the face sheets and the core. The theoretical predictions in the present work are compared with FEM results by ANSYS and the results published in the literature for special cases, and reasonable agreement is found between them.

Wrinkling and rupturing are major defects in forming of conical cups. Among the forming processes for producing conical parts with high drawing ratio in one stage, hydrodynamic deep drawing assisted by radial pressure (HDDRP) is a highly reliable method. This method has some advantages compared to conventional deep drawing such as more appropriate thickness distribution, better surface quality, higher limit drawing ratio, higher dimensional accuracy and ability of forming complex parts. In this paper, process window diagram of HDDRP on forming of conical parts was obtained considering several parameters including pressure path, initial thickness and material of the blank, by finite element method. FEM results were validated by experimental tests. The results show that the obtained process window diagram can predict appropriate forming area, and possibility of wrinkling and rupturing occurance with different pressure paths. Comparison of Simulations with experimental parts shows suitable conformity. The results can provide an applicable guidlines for sound hydroforming of conical products in automotive and aerospace industries.

Using a dynamometer is a test procedure for electric motors, combustion engines and gas turbines in desired and controlled conditions. In gas turbine, performance assessment and efficiency assurance is to be done by a dynamometer. In this paper, the flow around a rotating disk which is located at a given distance of the stator is simulated using computational fluid dynamicsmethods and comparison between the obtained results and the available experimental data shows good agreement.Distribution of pressure, temperature, flow rate between the rotating disk and the stator, torque and dimensionless torque coefficient for the rotational speed of the rotating disk, the gap between the rotating disk and the stator, shrouded and unshrouded, inlet nozzle diameter and the net radial inlet nozzle will be reviewed. The results show that dimensionless torque coefficientincreased by reducing the clearance between the disk and the stator or increasing shroud on the stator output but it didnt make a significant change byincreasing distance between the disk and the stator. The heat transfer rate and torque are also increase by diskrotational speed increment.

According to the importance of tank vehicles in carrying liquid and fuel and also low levels of rollover threshold, studying effects of baffles on the stability and dynamic behavior of the vehicles are very important. In this paper, in order to investigate the lateral acceleration of the tank vehicle as well as the transient response, the mapping method is applied to convert circular cross section into rectangular strip. The mechanical spring-mass model is used to analyse the sloshing model. Then, the the modal masses are obtained. Using the baffles decreases the free surface elevation as well as the forces and moments acting on the walls of the container. Horizontal baffles reduce lateral forces to a greater degree in comparison to vertical baffles. In horizontal baffle the maximum rollover threshold acceleration occurs in the 50% filling and baffle length of L/d=0.01 for two time t=1.95s and t=3.65s. and the value is constant in ay=0.439(g) for other times . Also in the surface-01 piercing vertical baffle, the maximum rollover threshold acceleration of the 50% filling and length of baffle L/d=0.99 is ay=0.45(g).

Continuous hydrothermal flow synthesis processes are implemented for the manufacture of nanoparticle metal oxide. In this processes, metal oxides temperature should be reduced with a convenient rate to avoid forming of larger crystals, which may affect the quality of the products. Using a high-efficiency heat exchanger can help us to gain this purpose. In this study, the first step is to model the heat exchanger in CHFS (Continuous hydrothermal flow synthesis) system numerically, in order to reduce the temperature of the nano-metal oxides. The numerical results are also validated with available experimental data. Then, in the next step, further reduction of product’s temperature is obtained by using a porous media, which eventually leads to an improvement in the products quality. Using the porous media increases the reduction rate of temperature for about 40%, that means 20 oC/s more than the typical heat exchenger and also decreases the required length of the heat exchanger about 35%.

Analysis of reciprocating flow is significant due to the different nature of this flow. An application of reciprocating flow is in heat exchanger of Stirling engine. The most of researches has been often considered to be laminar flow and incompressible, that it isnt appropriate for Stirling engines. In this study, a three-dimensional numerical simulation of reciprocating flow in Stirling engine heat exchanger in a wide of range non-dimensional flow displacement (20-100), high frequency of engine (30-130) and work pressure (5.25-11.5bar) were performed. The flow was considered compressible and turbulent. The characteristic of heat transfer in heat exchanger, the effect of oscillation frequency variation, working pressure and working fluid was studied. It was found that increasing the kinetic Reynolds number, working pressure and non-dimensional flow displacement increase heat transfer in engine's heat exchanger.In the ST500 Stirling engine was observed to increase of 80 percent of non-dimensional flow displacement, oscillation frequency and working pressure enhance 14, 9 and 20 percent the Nusselt number respectively. By replacing the hydrogen instead of helium, a 48 percent increase in heat transfer was observed.

Due to the importance of the role of turbines in various industries, a more precise design is needed to improve their efficiency. In this line, a more exact solution of the flow passing through turbine blades is needed. Nowadays, the advancement of numerical modeling techniques has resulted in achieving more accurate tools capable of considering the discontinuities with less oscillation and numerical errors. In this study, for modeling the flow in a supersonic 1D converging-diverging channel, the AUSM upwind method and a computational standard grid are used. The obtained results for a strong vertical shock agrees well with analytical results, and is considered as an initial verification of the method. In this research and for the first time, the AUSM numerical method has been further developed by using a simple H-grid mesh for modeling an inviscid subsonic and supersonic outlet flow in turbine stator blades. Comparing the model results with the results of Jameson’s artificial dissipation technique and experimental data showed that both numerical methods have good agreement with experimental pressure data, but the AUSM method has less numerical errors and improves the mass conservation by at least 25%. It is to be noted that the AUSM method is highly recommended for higher Mach numbers.

It is expected that the ordinary heat and power production systems in residential section are substituted by cogeneration systems in near futur due to higher overall efficiency. Between different cogeneration systems, fuel cell based systems are a suitable choice due to high efficiency, high power density, low emission and noise. In this paper a cogeneration system based on solid oxide fuel cell were examined on energy and exergy basis at first, then using optimization algorithms and a choice of three goal function electric power generation, heat production and minimizing waste Exergy, the operation of the system were optimized. The results included the calculating the working parameters of system with three goal functions and show that the most change of flow exergy is in fuel channel of fuel cell stack and most irreversibilities are due to recovery (38%), catalyst burner (37%) and fuel cell (16%). Among the internal components, air compressor is the biggest power consumer, with (14%) of produced power. Optimization results also show that the minimum exergy destruction is in the electricity production approach, thus using a CHP system is more preferable rather than the single system for producing power or heat.

In the current study, force convection heat transfer from a stationray circular cylinder in non-newtonian fluids has been simulated using the lattice Boltzmann method in 2D and unsteady flow regime. The simulations were performed for three Reynolds numbers (80, 100 and 120), with Prandtl numbers 10 and 20, the different non-newotonian power-law index in the range of 0.4 to 1.8 while varying the Brikman number from 0, 1 to 3. The effects of different Parameters on the vorticity contour, isotherm patterns, local Nusselt number and periodic-surface average Nusselt number is studied. The results show that viscous dissipation plays main when the operating fluid is non-newtonian. As the fluid behavior changes from shear-thinning fluids to Newtonian and then to shear-thickening fluids, the length and width of the wakes increase and sepration between the isotherms contours deacreases. Broadly, the rate of heat transfer deacreses with decreasing Reynolds and Prandtl numbers, and with increasing value of the Brikman number and the power-law index.

In this paper the POD method has been used for development of a reduced order model of the unsteady incompressible flow fields. After projection of the governing equations along POD modes, a low-dimensional dynamical system is achieved. Due to variations in energy level of POD modes and large differencing in modes’ energy, the outcome dynamical system can be converted to a reduced order model with a few amount of error. The standard low-dimensional models, due to some reasons, do not predict time variations of flow field accurately. One of these reasons is due to non-divergence free modes. Accuracy of dynamical system has been improved by using combined form of continuity and linear momentum equations which has been proposed by the author. The obtained reduced order model can predict time variations of flow field with fast computational speed and relatively good accuracy. The results from this low-dimensional model have been compared with direct numerical simulation data and shown good accuracy and salient simplicity in computations.

Numerical analysis and simulation of industrial lubricants in journal bearings are very important because of their numerous applications in various industries such as power plants, turbomachinery, electrical machinery, shipbuilding, and etc. In these investigations, valuable information such as temperature distribution of pads and oil, heat and friction losses, load capacity, and etc. are extracted which are used by the designers and constructors to improve the performance of bearings. In this paper, a numerical three-dimensional thermo-hydrodynamic code has been developed to simulate the steady condition of tilting-pad journal bearings without restrictions on their size, especially length of bearings. In this program, Reynolds equations for oil flow in the gap between the shaft and the bearing pads are solved by using a numerical finite difference method with a successive over-relaxation scheme. In this simulation, for closing the solution to the real conditions, oil viscosity changes with temperature and the deformation of the pads is also taken into account. To assess the effect of the physical properties of oil-bearing on hydrodynamic behavior of bearings, several important and widely used industrial bearing oils have been selected and the results are presented in this paper. Friction loss, the maximum temperature of the pads, oil flow rate, the minimum thickness of the oil Film and the pads tilting angle are the main presented results.

In this study, the numerical analysis of turbulent flow and heat transfer of oscillating impinging slot jet on an asymmetric concave surface has been investigated. In this way, the averaged Navier-Stokes equations for turbulent incompressible flow in an unsteady state with k-ε RNG turbulence model and in 2D computational space were solved. The effects of the oscillation frequency, period of oscillation, curved surface, jet distance to surface and jet Reynolds number on time- averaged Nusselt number distribution in the concave surface were studied. The obtained results show that applying the pulsating jet in the range of 40 Hz to 160 Hz can increase heat transfer from the concave surface in comparison with the steady jet. Furthermore, increasing Re Number from 4000 to 8000, and period of oscillation from 0.4 to 1, leads to the increase of the time-averaged Nusselt number. While that with increasing of distance between jet and concave surface the time- averaged Nusselt number, decreased.

A characteristic-based approach is developed for thermo-flow with finite volume methodology (FVM) in which multidimensional characteristic (MC) scheme is applied for convective fluxes. To obtain compatibility equations and pseudo characteristics, energy equation is taken into account in the MC scheme. With this inherent upwinding technique for evaluating convective fluxes at cell interfaces, no artificial viscosity is required even at high Reynolds numbers. As benchmarks, forced convection between parallel plates and forced and mixed convection in a cavity are examined for a wide range of Reynolds, Grashof and Prandtl numbers. Results confirm the robustness of MC scheme in terms of accuracy and convergence. The results obtained by new proposed scheme are in good agreement with the standard benchmark solutions in the literature. For time descritization, fifth order Rung-Kutta and for viscous terms, the secondary cells are used. Obtained results by new characteristic method were compared by other results that were in literatures.

Since the Robot dynamic equations contains nonlinear parameters, nonlinear control techniques could be a logical choice to achieve proper tracking and asymptotic convergence. According to this principle, in this paper, backstepping control method is applied to control a SCARA robot at the presence of uncertainty. The backstepping control method is a systematic technique for designing stable nonlinear control systems based on the Lyapunov stability criterion. In practice, the proposed controller should be robust to overcome the disturbances and uncertainties. Therefore in this study, using integrated back-stepping method, a stable controller of SCARA robot is designed with the aim of tracking trajectory. With respect to that the mass and inertia of the robot are not determined practically, an adaptive backstepping controller is implemented to estimate unknown parameters of the system. These controllers are simulated considering SCARA robot with three rotational and one prismatic joint. The results of backstepping method in the presence of external disturbances, and without it, reveals that the backstepping method is a right choice for both cases.