Sharif Digital Repository

This paper presents three schemes of 2D meshless finite volume (MFV) method, referred to as MFV with overlapping control volumes (MFV1), MFV with irregular non-overlapping control volumes (MFV2) and MFV with regular non-overlapping control volumes (MFV3). The methods utilize the local symmetric weak form of system equation and the interpolation functions constructed using the weighted multi-triangles method (WMTM) which is recently developed by the present authors. The proposed formulation involves only integrals over the boundaries of control volumes. The performance of the proposed schemes is studied in three benchmark problems. A comparative study between the predictions of the above MFV... 

In order to be a powerful tool, finite-element and finite-volume methods must be capable of handling complex flow in complex geometries. In this work, a structured finite volume element method is suitably developed for solving incompressible flow on a collocated grid topology. The method is generally-applicable to arbitrarily shaped elements and orientations and, thus, challenges the potential to unify many of the different grid topologies into a single formulation. The correct estimation of the convec-tive and diffusive flux terms at cell faces remarkably enhances the solution accuracy of the extended formulation. It is shown that the current formulation is enough robust to treat any... 

In this work, a finite-volume-based finite-element method is suitably developed for solving incompressible flow and heat transfer on collocated hybrid grid topologies. The method is generally applicable to arbitrarily shaped elements and orientations and, thus, challenges the potential to unify many of the different grid topologies into a single formulation. The key point in this formulation is the correct estimation of the convective and diffusive fluxes at the cell faces using a novel physical influence scheme. This scheme remarkably enhances the achieved solution accuracy. It is shown that the extended formulation is robust enough to treat any combination of multiblock meshes with dual... 

This paper presents an extension of the multiscale finite volume (MsFV) method to multi-resolution coarse grid solvers for single phase incompressible flows. To achieve this, a grid one level coarser than the coarse grids used in the MsFV method is constructed and the local problems are redefined to compute the basis and correction functions associated with this new grid. To construct the coarse-scale pressure equations, the coarse-scale transmissibility coefficients are calculated using a new multi-point flux approximation (MPFA) method. The estimated coarse-scale pressures are utilized to compute the multiscale pressure solution. Finally a reconstruction step is performed to produce a... 

A new approach has been presented to solve gas-liquid flow numerically in vertical pipes of air-lift pumps. To improve modelling, a new strategy has been employed with the capability of coupling the continuity and momentum equations and enforcing the role of pressure directly in the continuity equation. This is achieved via applying a novel scheme called the Physical Influence Scheme (PIS). The current finite volume solution is compared with other available numerical solutions. Indeed, they are in fair agreement. However, the present predictions are far superior to those obtained from an existing simple method, which is widely used in airlift pump modelling  

The numerical modeling of natural convection fluid flow and heat transfer in a quarter of gearwheel-shaped heat exchanger is carried out. The heat exchanger is included with internal active square bodies. These bodies have hot and cold temperatures with different thermal arrangements. Three different thermal arrangements are considered and showed with Case A, Case B and Case C. The CuO-water nanofluid is selected as operating fluid. The Koo-Kleinstreuer-Li (KKL) correlation is utilized to estimate the dynamic viscosity and thermal conductivity. In addition, the shapes of nanoparticles are taken account in the analysis. The Rayleigh number, nanoparticle concentration and thermal arrangements... 

The main purpose of this work is to improve the efficiency and performance of a primitive finite volume element method which provides superior capability on a single platform. This method is suitably extended in order to use the advantages of parallel computing on multiprocessors or multicomputers. The method is fully implicit which renders huge sparse linear algebraic kernels. Nevertheless, the attention is focused on solving the sparse system rather than constructing it. The current method is a cell-centered scheme. Since the grid is unstructured, each non-boundary node engages with nodes on three or more surrounding elements around that node. Depending on global node numbering, the... 

Interpreting the complex interaction of nanostructured fluid flow with a dipole in a duct, with peripherally uniform temperature distribution, is the main focus of the current work. This paper also sheds light on the changes in the Nusselt number, temperature profiles, and velocity distributions for the fully developed nanofluid flow in a vertical rectangular duct due to a dipole placed near a corner of the duct. A finite volume approach has been incorporated for the numerical study of the problem. It is interesting to note the unusually lower values of the Nusselt number for the higher values of the ratio Gr/Re. Due to the nanostructure in the fluid, an enhancement in the Nusselt number has... 

In the present work, the multiscale finite volume (MsFV) method is implemented on a new coarse grids arrangement. Like grids used in the MsFV methods, the new grid arrangement consists of both coarse and dual coarse grids but here each coarse block in the MsFV method is a dual coarse block and vice versa. Due to using the altered coarse grids, implementation, computational cost, and the reconstruction step differ from the original version of MsFV method. Two reconstruction procedures are proposed and their performances are compared with each other. For a wide range of 2-D and 3-D problem sizes and coarsening ratios, the computational costs of the MsFV methods are investigated. Furthermore, a... 

Multiscale finite volume (MSFV) method have been developed and applied in various complicated physics. The most important advantage of MSFV method is its computational efficiency. In this paper we present a new set of boundary condition for calculation of basis and correction functions which leads to further reduction in computational time in problems with medium heterogeneity and therefore improves computational efficiency. In standard MSFV (sMSFV) method reduced boundary condition is used to determine the basis and correction functions which is based on local information, however in modified MSFV (mMSFV) method global information is used at initial time for constructing boundary condition... 

Highlights: The objective of this work is to investigate the effects of the surfactants on the oil extraction from the dead ends through the numerical simulations and experimental evidences. The volume of fluid approach in the frame of the finite volume method has been used for numerical simulations in 2-D domain and experimental flooding tests have been done using a glassy micro-model. The effects of the water-oil, water-wall and oil-wall interfacial tensions have been investigated numerically and some results are compared to experimental flooding results. Simulations have been done in the cases of water-wet, neutralized-wet and oil-wet micro-models also. The numerical results show that in... 

In this work, basic design features of Rayan are documented. One of the new design features presented in this work is the way Rayan handles polyhedral grids. Grid definition is combined with the definition of the structure of the sparse coefficient matrix, thereby releasing a considerable part of the memory used by the grid to store otherwise required faces belonging to the cell part of the connectivity description. The key idea is to use a uniform way for creating the structure of the coefficient matrix from the grid connectivity description and to access that data when computing the elements of the coefficient matrix. This saving requires many modifications to the computational algorithm... 

The main objective of the current work is to introduce a new conceptual linearization strategy to improve the performance of a primitive shock-capturing pressure-based finite-volume method. To avoid a spurious oscillatory solution in the chosen collocated grids, both the primitive and extended methods utilize two convecting and convected momentum expressions at each cell face. The expressions are obtained via a physical-based discretization of two inclusive statements, which are constructed via a novel incorporation of the continuity and momentum governing equations. These two expressions in turn provide a strong coupling among the Euler conservative statements. Contrary to the primitive... 

Cavitation phenomenon is defined as the process of rupturing any liquid by a decrease in pressure at nearly constant temperature. The cavities driven by the flow in a region of high pressure will implode and generate high pressure pulses leading eventually to erosion and vibration. But in supercavitation the bubbles produced by cavitation combine to form a large, stable bubble region around the supercavitating object. This phenomenon decreases the drag on the supercavitating body. Experimental testsware performed at 2-D unsteady flow for two wedge shaped bodies made before in laboratory and cavitation inception and its development were captured by a high speed camera. Then this cavitation... 

The rapid progress in fabricating and utilizing micro-electromechanical systems during the last decade has not been matched by the corresponding advances in our understanding from the unconventional physics involved in manufacturing and operation of micro devices. To avoid the complexity encountered in modeling of nonlinear Boltzmann equations, the Navier-Stokes equations can be solved considering the slip flow regime concepts. The modeling can be achieved via employing suitable slip velocity boundary conditions at the solid walls. The modified first-order slip models can, in some cases, extend the range of applicability of the Navier-Stokes solvers to around and beyond Kn=0.1, where the... 

The dynamics of a two dimensional plane jet injected at the base of a step, parallel to the wall, in backward facing step flow geometry is studied. The objective of this work is to gain insight into the dynamics of the igniter flow field in solid fuel ramjets (SFRJ). The one equation turbulence model of Spalart and Allmaras is used. The system of governing equations is solved with a finite volume method using a structured grid in which the AUSM+ scheme is used to calculate the convective fluxes. It is shown that the wall jet drastically changes the structure of recirculating region of back-step flow. The wall shear stress caused by the wall jet is much greater than that caused by the main... 

A new procedure for simulating turbulent flow in three-dimensional arbitrary geometries is presented. Finite volume method using physical covariant velocities on a staggered grid arrangement was used in this investigation. This work is an extension of previous successful work to three-dimensional cases. The ability of the new algorithm was tested using a conventional two-equation turbulence model on a highly separated turbulent flow test case. The low Reynolds number k-ω turbulence model of Wilcox was utilized to evaluate its capability in modeling highly curved flows. Turbulent flow over a three-dimensional hill, which is appropriate in assessment of ability of turbulence models in... 

In two phase flow investigation, there is a need for robust methods capable of predicting interfaces, in addition to treating the traditional governing equations of fluid mechanics (Navier-Stokes Eqs.). Such methods in a finite volume approach can be classified into two typical categories called interface tracking and interface capturing methods. According to their abilities, interface capturing methods are of more interest in free surface modeling, especially when complex interface topologies such as wave breaking are included. These methods solve a scalar transport equation in order to find the distribution of two phases all over the computational domain. That is, all properties of the... 

A numerical method based on the SIMPLE algorithm has been developed for the analysis of vapor flow in a concentric annular heat pipe. The steady-state response of a concentric annular heat pipe to various heat fluxes in the evaporator and condenser sections are studied. The fluid flow and heat transfer in the annular vapor space are simulated using Navier-Stokes equations. The governing equations are solved numerically, using finite volume approach. The vapor pressure and temperature distributions along a concentric annular heat pipe are predicted for a number of symmetric test cases. The vapor flow reversal and transition to turbulence phenomena are also predicted. The results are compared...