مجله مهندسی مکانیک امیرکبیر
سال پنجاه و سوم شماره 7 (مهر 1400)

With the advent of morphing airfoils, the aerodynamics of wind turbines and wings underwent many changes. In this study, the aerodynamic coefficients of morphing airfoil based on NACA 0015 are optimized in the range of Reynolds number 105 to 106 and the angle of attack 0 to 12 degrees using Artificial Neural Network (ANN) and Genetic Algorithm(GA). First, the airfoils were created in MATLAB software by random control points and mesh generated in Gambit software, then in two-dimensional Ansys software were simulated. The simulation results, including lift and drag coefficients, separation point and pressure center, with control points were used to train the Artificial Neural Network (ANN). The trained function is given as an input function to the Genetic Algorithm(GA) to optimize the desired coefficients.Lift coefficient, the center of pressure, separation point and lift to drag ratio were optimized as a single objective, In single-objective optimization, the lift coefficient was increased by 18% using the morphing airfoil. Also, the lift coefficient and the center of pressure, the lift coefficient and the drag coefficient were optimized as the dual-objectives optimization. In the optimization of the dual objectives, lift and drag coefficients were controlled by 0.8 and 0.03, respectively, by the morphing airfoils.

In this paper for flow simulation in fractured porous media, a multiscale finite volume method on unstructured grids is developed. To this end, algorithms for generating coarse scale unstructured grids for the matrix and fracture networks are presented independently. The presented algorithms for generating coarse scale unstructured grids are adaptable based on local changes in permeability field. Unstructured grid adaption based on permeability field has significant effect on improving the multiscale solution results in highly heterogeneous permeability fields. For the first time in this research, applying adaptive unstructured grids in fractured porous media is done. Coarse scale grid cells are generated such that strong variation of permeability along their boundaries and also the placement of coarse scale vertices in low permeability region are prevented. To reduce the computational cost, fracture-matrix coupling is considered only for the calculation of basis functions in the matrix domain. In order to evaluate the proposed algorithms, various 2D test cases are designed and solved. Finally, it is shown that the multiscale finite volume method with the proposed algorithms is an efficient numerical method for flow simulation in heterogeneous fractured porous media.

When an axially symmetrical fluid jet impacts on a horizontal target plate vertically, a hydraulic jump is formed. Flow rate, fluid jet diameter, nozzle opening distance to the target plate, surface tension of the fluid, etc. are different parameters important for investigating this phenomenon. The impact of the important and key parameter of fluid jet swirl on hydraulic jumps is not investigated. The main purpose of this study is to investigate the impact of fluid jet swirl on circular hydraulic jump. The results of this study, achieved by using experimental method, show that the higher the angular velocity, the higher the increase in the radius of circular jump. The increase in jump radius with angular velocity of nozzles 132.8, 266.4 and 400.8 rpm compared to nozzle without swirl is 3.29, 5.89 and 8.34 percent, respectively. Drawing the diagram of the dimensionless radius of jump based on the dimensionless number of swirl shows two categories of lines. The first category is the fixed angular velocity lines with a negative slope and the second category is the fixed flow rate lines with a positive slope. Experiments also show that the hydraulic jumps created by a swirling jet follow the trend results of modified Watson's theory with a non-significant difference.

In this paper, the water hammer resulting from the fast closure of the valve in a pipeline is simulated using numerical solution of continuity and Navier-Stokes equations. Simulation has been performed for a high-viscosity oil and for water. The initial flow regime for water is turbulent and for the oil is laminar. The obtained results are compared with the experimental results and for both fluids, a good agreement between the simulation and experimental results at different times is obtained. Velocity contours at different times show two regions with different behavior in transient flow. The wall region and the pipe core region. In the wall region, the effects of fluid viscosity are dominant, the velocity gradients are sharper, and flow changes more rapidly. While the pipe core region is affected by fluid inertial forces. As the fluid viscosity decreases and the Reynolds number increases, the core region becomes larger. In addition, a parametric study has been conducted and the effect of different parameters on water hammer has been studied. The results show that by reducing the thickness or length of the pipe, reducing the mass flow rate, or using a pipe with a lower elastic modulus, the water hammer effects can be significantly reduced. For example, by reducing the length of the pipe from 60 meters to 18 meters, the maximum pressure decreases by 11%.

This paper reports on a numerical simulation of circumferential groove casing treatment in a high-speed axial fan. Four circumferential grooves of the same geometry are located over the tip of a NASA Rotor-67 and unsteady calculations are performed from choke to near-stall. Results show that circumferential grooves reduce the incidence angle near the pressure surface at the blade leading-edge. Furthermore, the passage shock and the leakage flow are pushed rearward in the passage. It is also found that circumferential grooves have two simultaneous important effects on the endwall flow field. First is adding momentum in the streamwise direction (fluid is absorbed by the grooves from their downstream part and is injected from their upstream section). The other effect is providing a flow path between the suction and pressure surface, leading to a reduction in the pressure difference between them (fluid is absorbed by the grooves mostly near the pressure surface and is injected into the passage near the suction surface). At the near-stall point the flow field near the grooves was found to be highly unsteady. Maximum unsteadiness was observed in the first upstream groove: the circulated mass flow rate changed as high as roughly 30 percent of its time-averaged value.

Clear air turbulence (CAT) forecasting got its start in the World War II era, when scientists and individuals in the field of aviation first began attempting to correlate observed clear air turbulence events with large scale synoptic features. Although there is a large spectrum of eddy sizes in the atmosphere, from macro scale to micro scale, aircraft bumpiness is felt mainly for a range of eddy sizes between about 100 m and 1 km; larger eddies cause only slow variations in the flight path while the effect of very small eddies is integrated over the surface of the aircraft.This study is two case study and comparison of the report of CAT of center, east and south east crossings (Sector 4, 5 and 6) of the results of the IRIAF IT, Communication and Electronic Administration. Regional weather predictions are carried out using an ensemble forecasting system development for the WRF model. Perturbed initial conditions and different choices of physical parameterizations are used to generate ensemble members. In addition, the initial and lateral boundary conditions are taken from the global forecast system (GFS). For each member of the ensemble system, two computational domains of the nesting area are used with spatial resolutions of 27,000 meters, 9,000 meters. Case study of the predicted clear air turbulence indicates the proper performance of the predicted meteorological parameters.

To date, due to growing investment in renewable energies, researchers are interested in wind power feasibility studies for mountainous terrains. Numerical Weather Prediction (NWP) methods has proved to be a strong tool to simulate wind flow field in real time. In present study, using NWP method and in particular WRF (weather research and forecasting model) software, the wind flow, over Martigny mountainous area located in Switzerland, has been simulated and studied. Due to high resolution of the simulation, Large Eddy Simulation (LES) has been employed to perform turbulence modelling. The objective of this study is to assess the credibility of WRF in wind simulation and to examine the effect of horizontal and vertical resolutions as well as two different sub-grid scale turbulence models. The results reveal that WRF is able to generate the wind flow in comparison with those data obtained from wind measurement stations and it can properly produce diurnal wind patterns. The results also show a promising simulation for the region, located within a wide and flat valley. However, the discrepancies between the results and those obtained from the wind station, are bold for regions located at high mountainous locations and at the peaks. Smagorinsky 3d SGS turbulence model show a better performance at mountain peaks in comparison with TKE1.5 model. In addition, increasing vertical resolution reduces the discrepancy and improves WRF performance.

Numerical modeling and simulation is a useful tool for analyzing the dynamic behavior of engineering systems. Using these methods, especially for unsteady problems, usually requires a lot of time. For this reason, the development of fast speed algorithms and increased computational efficiency has always been an important issue for researchers. In this method, by decreasing the constraints of the system, the computational speed will dramatically increase without changing the dominant features of the problems. In this research, using the basic concepts of dynamical systems, two problems such as diffusion and convection-diffusion of equations are investigated independently and by using the proper orthogonal decomposition method, the reduced-order model for these equations have been created. Finally, for each of the problems, based on the projection of the related governing equations in the vector space of modes and using more energetic modes, a reduced-order model is obtained with respect to the orthogonal basis functions. The model obtained has been used to simulate the time evolutions of each problem. These equations can correctly replace with the primary equation and predict the behavior of the system with very good accuracy. Comparison of the results of the reduced-order model and direct numerical simulation shows high accuracy.

Gravitational currents are important currents in atmospheric and oceanic studies. Gravity current is caused when a fluid with different density moves into another fluid. If the fluid of a given density enters the stratified ambient, such that its density is lower than the underneath layers and higher than the upper layers, the gravity current is of the intrusive type. The Kelvin-Helmholtz and the Holmboe instabilities are seen in the interface. The decisive parameters in the type of instability are the Richardson number local (J) and the ratio of shear layer thickness to the density layer (R). In this study, two-dimensional numerical simulation of the shear instabilities (Holmboe waves) with the Eulerian-Eulerian approach on intrusive gravitational flow is investigated. OpenFOAM code was used to perform this simulation, and due to the turbulence of the flow, the LES method was used to model the turbulence. The obtained results show that with increasing the intrusive current density, the value of Richardson number decreases and the R parameter increases. Also, as the density increases, the frequency of the Holmboe waves first increases, then decreases. An increase in the wavelength of Holmboe waves is observed with increasing the intrusive current density but in the wave number of the waves, the opposite behavior is seen. The phase velocity of Holmboe waves also does not have a specific trend with density changes.

Droplet generation with controllable precise size is the target of this research. For this purpose, a flow focusing micro-channel is constructed using photolithography. Then, by injecting two incompatible fluids to the channel, micro-droplets are generated. Two syringe pumps are used. The Meros high speed camera is used for recording the image of droplets, and a fast image processing algorithm is used to calculate the size of the droplets. The feedback loop is completed by connecting the camera, microscope, syringe pumps and PC computer. To regulate the size of the droplet, the PID controller is used due to its ease of implementation and robustness. The flow rate of the continuous phase is the control input and the size of the droplets is the output of the closed-loop system. Experimental tests are done by considering three desired droplet diameter, i.e. 100, 140 and 160 μm. In all cases, the controller shows good performance. To show the disturbance rejection characteristic of the designed system, the flow rate of the discrete phase is changed stepwise. Due to this disturbance the transient response of the system changed, but the controller attenuates this disturbance, and regulates the system to the desired size. The experimental tests show that the designed closed-loop microfluidic system is able to generate droplets with desired precise size.

There are several ways to increase heat transfer in mini-channels, like adding a vortex generator. In this paper, the effect of the presence of the hole on vortex generators on the heat transfer parameters is examined. In this study, a 50 mm long mini channel with eleven rectangular vortex generators, with holes with area of 5 to 60% of the vortex generator area, was analyzed with water-based fluid under constant flux in the range of Reynolds numbers 200-1000. The results showed the presence of holes on the vortex generators reduced the pressure drop resulting from the obstruction against the fluid flow and 34.7% decrease in pressure drop is observed for minimum and maximum area of holes in Reynolds number 1000. The Nusselt number is increased by existence of a hole in the Reynolds numbers range and then is decreased by increasing the size of the hole due to the reduction of the vortex size behind the obstacle, so that in the maximum Reynolds number by increasing hole size, 34.3% decrease is observed. The results also showed that the maximum performance evaluation criterion for the Reynolds number 200 is obtained in the VG-15 model and by increasing Reynolds number in the models with smaller holes, which maximum value of the performance evaluation criterion is occurred in VG-05 model in the Reynolds number 1000.

In this paper, the effect of baffle on the flow field and heat transfer enhancement of forced convection of Cu-water nanofluid flow in the laminar regime over a backward facing step is numerically investigated. In this study, the influence of baffle geometrical parameters as height, width and number of baffles, as well as the Reynolds number and the volume fraction of nanoparticles on the flow filed and heat transfer are evaluated. Also, to evaluate the simultaneous of the heat transfer enhancement and pressure drop, the Performance Evaluation Index (η) is calculated. The results indicated that in presence of baffle, the η is higher than 1 for all cases, and it is clear that the presence of baffle is effective on thermal efficiency. It is obvious that by increasing the Reynolds number and decreasing the volume fraction of nanoparticles, the Performance Evaluation Index is increased. For all cases, the average Nusselt number and the Performance Evaluation Index for wb=2H are higher than other cases about 7.6% and 15% respectively. The results show that using 2 baffles must be more beneficial than other number of baffles. Finally, a correlation for the Num/Num0 as a function of Reynolds number (Re), volume fraction of nanoparticles (φ), number of baffles (N), baffle height (hb) and baffle width (wb) is presented with an average error of 2.88%.

In this paper, flow characteristics and heat transfer in a smooth horizontal pipe subjected to forced heat convection with constant wall heat flux in the presence of magnetohydrodynamic has been computationally analyzed. The effects of temperature-dependent density, specific heat capacity, thermal conductivity, and viscosity on heat transfer and frictional flow characteristics of transformer oil and local and average heat transfer coefficient have been numerically investigated. Firstly, to validate, the present numerical result has been compared with the analytical and experimental results through a smooth pipe, which shows a good agreement. A significant deviation between constant and variable property has been achieved. Changes in fluid velocity profiles have led to changes in fluid characteristics including coefficient of friction and heat transfer coefficient. By considering the changes in the parameters, it was observed that the viscosity of the base fluid and the nanofluid have the maximum effect with approximately 30 and 25% increase in heat transfer coefficient and apparent friction coefficient relative to the fixed properties, respectively. Despite the dependence of the thermal properties of the nanofluid on temperature-dependent viscosity, the change in thermal conductivity leads to 35% increase in the heat transfer coefficient in the presence of a magnetic field.

In the current study, an experimental apparatus was constructed, and subcooled flow boiling in an annulus tube was investigated. The annulus tube is in the vertical direction, and the internal and external diameters are 50.7 and 70.6 mm. The operating pressure was 1 atm, and the working fluid was water. In this investigation, heat flux, mass flow rate and the inlet subcooling effectiveness on heat transfer coefficient are considered. The experiments were performed in the mass flow rate range of 0.012 kg/s to 0.0286 kg/s in which the flow consists of both forced convection and flow boiling. The results show that the heat transfer coefficient is highly dependent on heat flux in a direct relationship. The results revealed that the subcooled heat transfer coefficient increases with decreasing inlet subcooling. Also, the mass flow reduction causes heat transfer coefficient increments to 30% in subcooled boiling regions. The use of porous media also increases the subcooled flow boiling heat transfer coefficient up to 30%. The validation of empirical results has also been done with valid previous reports.

The dynamic variation of surrounding soil temperature in axial (depth) and radial directions of vertical type Ground Source Heat Pump (GSHP) heat exchangers are investigated here. This soil temperature distribution for borehole heat exchangers (BHE) plays an important role in thermal operation, electricity consumption and Coefficient of Performance (COP) of GSHP. Thus the transient 3D numerical modeling of U-tube and coaxial BHEs are investigated to find the temperature distribution around the buried pipes. The simulation is performed using ANSYS FLUENT 16.0 software based on the finite volume method. The effects of various parameters are studied and modeling results for the cooling application of heat pump are obtained for different mass flow rates of condenser cooling water. Results show that the injection heat transfer rate to the ground in summer, in the coaxial BHE with fluid injection direction to inner pipe at mass flow rates of 0.8, 1, 1.2 kg/s are 5.34%, 11.9%, 16.5% higher than U-tube BHE respectively. Moreover, after 93 days, the vertical temperature distribution of the soil for U-tube BHE shows a significant variation mainly at depths less than 36.6 meters while the vertical temperature variation with coaxial BHE is significant even in higher depths.

With the promoting use of lithium-ion batteries in industries, increasing their temperature is known as a challenge. In this research, by completing a hybrid heat management system, the effect of adding Nano-titanium oxide particles to paraffin PCM investigated experimentally on the cooling performance of battery in two constant heat flux, 4.5 and 14 watts. The hybrid system consists of nano-paraffin and copper metal foam, which has been used with two working fluids as air and pure water. Nanoparticles are added with pure paraffin in four volume percentages 1, 2, 3 and 4. In case of air working fluid, battery temperature in pure paraffin and nano-paraffins 1 to 4% became 56.2 °C, 51.8 °C, 50.7 °C, 49.3 °C and 48 °C, respectively. Thus. in case of nano-paraffin 4% comparing to the pure paraffin, temperature decreasing was about 17%. Hybrid system with pure water as a working fluid and using copper foam and nano-paraffin tested in Reynolds numbers 420, 600 and 720, and it is shown that the battery temperature reached to a stable temperatures of 48°C, 46°C and 44°C respectively, which comparing to the pure paraffin case, temperatures reduced by 11%, 12% and 12.5% respectively. Therefore, due to the low thermal conductivity of paraffin, the addition of nanoparticles in both working fluid cases of pure water and air is beneficial.

In this research, it has been tried to experimentally investigate the the effect of personalized ventilation system on air flow and temperature distribution in a room for two inlet air temperatures (24 and 32°C) and two different arrangements of the inlet diffusers (desk-mounted and under-desk air terminals). The results showed that the penetration depth of heat and momentum due to the inlet diffuseres were not the same; So that the effect of inlet diffusers’ temperature on the room temperature distribution is significant up to a distance of about 60 cm. However, the effect of inlet velocity is noticeable up to a distance of about 110 cm. As a result, the occupants’ thermal sensations up to about one meter from the inlet diffusers will be affected by the inlet conditions. Also, the results indicated that the draught dissatisfaction along with the diffusers’ centerline is significant to a distance of about 180 cm. Also, based on the results, the air velocity and turbulence intensity are the two main factors in determining the draught dissatisfaction in the personalized ventilation system and due to the rapid thermal mixing of inlet air with the room air, the effect of inlet temperature on the draught dissatisfaction is not significant.

Ventilation is essential to provide a smoke-free path for safe passenger evacuation and effective rescue services in case of a tunnel fire, because the closure tunnels increase consequences of accidents significantly. In the present study, the simultaneous use of longitudinal ventilation and smoke extraction from the ceiling in fires inside tunnels and physical phenomena has been investigated. Fire dynamics simulator will be used as a CFD tool. This simulation was performed to investigate the effect of the longitudinal distance of the smoke extraction system from the fire source on the smoke back-layering length and the maximum temperature in the two operating conditions used by this system downstream and upstream the fire source. In the present work, the smoke extraction system is located on the ceiling of the tunnel. The results show that using a smoke extraction system upstream the fire source will increase the maximum temperature, but using the same system downstream will reduce the temperature throughout the tunnel and prevent smoke back-layering. However, attention to the smoke extraction velocity for prevent the plug-holding phenomenon. The results also show that the simultaneous use of two smoke extraction systems at the upstream or downstream of the fire source will have a better result and The maximum temperature is reduced by 10%.

The photovoltaic/thermal system is capable of generating both heat and electricity simultaneously. The purpose of using spectral filters is to make full use of the solar radiation spectrum and thermal separation of photovoltaic and thermal units. The purpose of this paper was to investigate a new hybrid spectral filter consisting of phase change material and nanofluid in order to achieve a filter close to the ideal spectral filter. In this regard, the photovoltaic/thermal system with a combined nanofluid-phase change material spectral filter was simulated using energy balance equations in MATLAB software and its performance was compared with two conventional and nanofluid-based spectral splitting photovoltaic/thermal systems from energy and exergy viewpoints. Also, the optical properties of nanofluid and phase change material were simulated and the models were validated with the experimental data available in the literature. The results showed that by using a combined filter the photovoltaic temperature can be reduced by up to 50% and the output fluid temperature can be increased by twice. The exergy efficiency of the system with the combined filter was about 14% and 22% higher and its exergy destruction was about 5% and 7% lower than conventional and nanofluid-based spectral splitting photovoltaic/thermal systems, respectively. The system also achieved the highest exergy efficiency at concentration ratios greater than 15.

The Stirling engine has attracted researchers' attention in recent years due to some advantages such as low noise, external combustion, and the ability to use solar and other new energy sources. Moreover, these engines can also be used in applications with a low or high-temperature difference. The type of cylinders, their arrangement, and the transmission mechanism can affect this engine's performance. On the other hand, engineers and designers are always looking to increase the efficiency and effectiveness of mechanical systems, which in engines can lead to increasing the engine's work or power. In the current study, firstly, the dimensional analysis of different types of Stirling engines is done. Then, by defining the engine's geometric parameters as the design variables, the engine's output work will be maximized using optimization algorithms. Also, in order to prevent the increase of the dimensions of the engine and its occupied space, a new constraint in the problem will be used. Kinematic optimization is applied to four different types of Stirling engines. Three algorithms, namely GA, PSO, and ICA, have been used to solve the optimization problem. The results of kinematic optimization show that the output work of the engine with optimal dimensions has increased approximately 1.45 to 4.59 times.