Papers by Keyword: Finite Volume Method (FVM)

Paper TitlePage

Authors: Mehdi Farrokhnejad, Anthony Straatman, Jeffrey Wood
Abstract: Solidification of a wedge casting of magnesium alloy AZ31 in the presence of melt convection is investigated. The formulation used in the in-house code is developed based on the finite-volume method on unstructured grids and utilizes a colocated finite-volume framework. In the discretization of the mass and momentum equations all novelties proposed in [1] are implemented and a source based-energy equation is presented. The volume fraction of solid in the domain and the cooling rates at six different embedded thermocouples are predicted. These results are compared to previous experimental work and against similar results obtained with a widely used commercial software package.
Authors: Jun Qiang Zhang
Abstract: The meshless finite volume method was employed to enforce the conservation of momentum in a local weak form. A novel hybrid meshless finite volume method (HMFVM) was proposed. The displacement and the stress were interpolated independently with a hybrid scheme in the process. To enforce the compatibility between stress and displacement, a smoothing stress was introduced via finite volume method. In this way, the HMFVM can avoid the appearance of derivatives of shape functions and improve the efficiency over the orthodox meshless finite volume method. Then, the HMFVM was applied to some Elasto-plastic problems, illustrating that it enjoys high precision and efficiency. As a result, the derivatives of shape function are avoid absolutely.
Authors: Feng Xiao, Bin Xie
Abstract: A robust and practical CFD code has been developed. The numerical framework, so-called VSIAM3 (Volume/Surface Integrated Average based Multi-Moment Method) makes use of two kinds of integrated moments of physical field, i.e. the volume integrated average (VIA) and the surface integrated average (SIA), which are treated as the computational variables and separately updated in time. VSIAM3 formulation is essentially different from conventional finite volume method and provides a convenient and robust framework to accommodate many existing numerical techniques for simulating various complex flows. In this paper, we will present the underlying idea of VSIAM3 and the extensions to make it applicable to various practical problems. Efforts toward high computational performance on hard wares with distributed memory and GPGPU will be also reported.
Authors: S. Torfi, S.M. Hosseini Nejad
Abstract: In this paper, a numerical model is developed to simulate single droplet heat and mass transfer in a two-pieces solution with a saturated solvent vapor environment. Finite volume method and transient SIMPLEC algorithm in spherical coordinates system used for simulation. For simulation of the mass transfer, dimensionless equations of motion, heat transfer and mass transfer (based on mass ratio) are solved simultaneously. All the thermodynamic and transitional solution properties have been considered as a function of temperature and concentration. Verification of method is done by compare these numerical results with analytical and numerical analysis of other studies. Evaporation, absorption and condensation contours in a water droplet at uniform temperature at superheated water vapor and Distribution of Droplet's Surface Temperature, Constant Temperature and concentration lines as the modeling results are presented. The results are shown that 0.5% increase of concentration of droplet cause increase 8 degree C of mean temperature of droplet.
Authors: Sudi Mungkasi
Abstract: We extend the application of numerical entropy production, as a smoothness indicator, from conservation laws to balance laws. We aim to indicate the smoothness of solutions to the shallow water equations involving varying width, which are a system of balance laws. The numerical entropy production appears to be accurate to detect discontinuities. As a numerical test, a radial dam break is considered. We assume that there is a higher level of water inside a radial dam than water outside the dam wall. If the radial dam is totally broken, then water flows from inside to outside. The flow results in a solution having shock discontinuities. Finding the positions of the discontinuities is our interest. They are the positions where numerical solutions, such as those generated by a finite volume method, decrease their accuracy. Detecting the position of the discontinuity can help in the improvement of the numerical solution in terms of its accuracy. We obtain that the numerical entropy production is simple to implement but give an accurate detection. The discontinuity of the stage (free water surface) is clearly detected by large values of the numerical entropy production as the smoothness indicator.
Authors: Fatimah Yusop, Bambang Basuno, Zamri Omar
Abstract: Computational fluid dynamics (CFD) is very widespread use every day as a tool in fluid flow analyses. In order to solve the Partial Differential Equation (PDE), there are few approach been introduced. The total variation diminishing (TVD) is a most popular scheme which is usually used in combination with other scheme. Therefore, this study develops CFD code by using Runge-Kutta which based on combination of central scheme and TVD scheme. Comparison was done through purely Runge-Kutta and after implemented TVD. The result shows that combination of Runge-Kutta and TVD approach are more stable as compared to purely Runge-Kutta approach.
Authors: Jong Kwan Hwang, Keun Yong Sohn, Kyung Hyun Kim, Dae Min Kang
Authors: Yan Li, Yong Teng Jing, Longnv Li
Abstract: The AutoCAD, Gambit and ANSYS software are used to establish transformer oil flow-temperature rise model and mesh generation based on finite volume method, fluid mechanics and numerical heat transfer. A method that calculates temperature rise distribution of transformer winding regional wire and oil by FLUENT software, and a numerical example is given for an actual transformer.
Authors: Ya Jun Wang
Abstract: A method is implemented to get the pressure distribution of the axial piston pump slipper. Slipper was seen as translating thrust bearing, taking slipper tilt and spin in account, based on finite volume method, hydrodynamic and hydrostatic pressure has been calculated by using the mass conservation principle. For a representative element volume, the difference flow was averaged by the difference flow between the tilting planes, and the shearing flow by slipper translating was averaged by the shearing flow between the tilting planes. The numerical calculating result based two liquid resistance assume was compared, the results showed that two methods have got the same pressure distribution schematics, and the high pressure area locates at the slipper titling direction, but for the pressure values at high pressure area, the second method is slightly higher than the first method, and that the higher pump speed were, the higher the pressure values, and at the same pump speed, the slipper spin speed affects slightly the pressure, and at the lower pump speed, the lubricant pressure tends to the hydrostatic lubrication.
Showing 1 to 10 of 101 Paper Titles