Cavitation in a valve leads to trouble and inconvenience for factories. Valves in a piping system are ruined, leading to costly replacements every several months. To reduce the cost caused by cavitation in a valve, a cage is utilized to make cavitation occur only in the region adjacent to the cage itself; therefore, only the cage needs to be replaced. To validate the design of a cage, simulation of the turbulent flow field inside a globe valve and the occurrence of cavitation are necessary for a valve designer. To reach this purpose, prediction of the cavitation inside the globe valve with and without a cage is undertaken, and a cavitation model is established in this study. The percentage of vapors in each computational cell is calculated using the proposed cavitation model. Two various cages, the one-stage perforated cage and the one-stage step cage, are considered. Vapor resulting from cavitation appears in the vortices existing inside the valve and at the downstream region of the globe valve w...
1 February 2013 in Journal of Hydraulic Engineering
A hydraulic jump is a common phenomenon that can be observed in open channel flows such as rivers and spillways. It can cause damage of the downstream bed and bank of the channel due to the process of continuous erosion and degradation. In order to reduce the hydraulic jump destruction, the energy in the hydraulic jump must be dissipated as much as possible. One method to increase dissipation of energy is using a corrugated bed. In order to know the effect of a corrugated bed on the hydraulic jump, a smoothed particle hydrodynamics (SPH) model is applied to investigate the characteristics of hydraulic jumps in various corrugated beds. A variety of corrugated beds that are smooth, triangular, trapezoidal, and sinusoidal are considered. The opening of a gate is changed to adjust the hydraulic jump. The conjugate depth ratio, the jump length, the bottom shear stress distribution, and the energy dissipation are reported. The results of the present study are in a good agreement with previous studies. E...
27 December 2012 in Computational and Mathematical Methods in Medicine
Pulmonary regurgitation (PR) is a common phenomenon in pulmonary arteries in patients after repair of tetralogy of Fallot (TOF). The regurgitation fraction of left pulmonary artery (LPA) is usually greater than right pulmonary artery (RPA) according to clinic data. It may be related to blood flow in pulmonary arteries. Therefore, understanding hemodynamics in pulmonary arteries helps to comprehend the reason. The aim of this study is to use 3D reconstructed pulmonary artery models from magnetic resonance imaging (MRI) and to use numerical approaches for simulation of flow variations in pulmonary arteries after repair of TOF. From the numerical results, the blood flow is influenced by the bifurcation angles and geometry of pulmonary artery. The regurgitation happens first in LPA after repair of TOF due to the small angle between LPA and main pulmonary artery (MPA). The recirculation region which obstructs forward blood flow to the left lung is found in LPA during acceleration of systole. We also analyze the pressure distribution; the extreme pressure variations are in dilation area of MPA. Numerical data including regurgitation in MPA, LPA, and RPA are compared with phase contrast MR measured data. Good agreements are found between numerical results and measured data.
The utilization of data in physics-informed neural network (PINN) may be considered as a necessity as it allows the simulation of more complex cases with a significantly lower computational cost. However, doing so would also make it prone to any issue with the data quality, including its noise. This study would primarily focus on developing a special loss function in the PINN to allow an effective utilization of noisy data. However, a study regarding the data location and amount was also conducted in order to allow a better data utilization in PINN. This study was conducted on a lid-driven cavity flow at Re = 200, 1000, and 5000 with a dataset of less than 100 velocity data and a maximum noise of 10% of the maximum velocity. The results show that by ensuring the data are distributed in a certain configuration, it has zero noise, and by using as much data as possible, the computational cost of PINN can be significantly reduced compared to without using any data at all. For Re = 200, it is 7.4 faster by using data, and this speedup is potentially higher for higher Re cases. For the noise in particular, it does not only make the PINN more inaccurate but also necessitate the usage of more data as this is the only way to make it more accurate. This issue though is capable to be solved with our new method, which only uses the data as an approximate solution, and the governing equation would figure out the details. This method was also shown to be capable to improve the PINN accuracy with the potential to almost completely eliminating the noise effect.
Numerical investigation of the effects of boundary conditions and computational domain size on simulations of a circular cylinder in three-dimensional (3-D) turbulent flow at has been conducted using a direct-forcing immersed boundary (DFIB) method. The turbulent flow scales were simulated using the Smagorinsky model. Two computational domain sizes and three different sets of boundary conditions were investigated. The results of all the cases were compared against published literature. The study proved the capability of the selected DFIB method for investigating fluid-structure interactions in turbulent flow. It also found that the increase in the size of the computational domain improved the accuracy of the results, especially in the far wake region. Interestingly, switching the lateral boundaries between periodic and symmetry conditions enabled the numerical simulation to approach the two different experimental results.