Papers by Author: Tao Lu

Abstract: Turbulence mixing in T-junctions filled with porous metal materials is numerical investigated using Large-eddy simulation turbulence model. Three cases of porous metal materials, made of three sintered metal spheres with different thermal conductivities (387.6 W/m۟۬•K for copper, 202.4 W/m۟۬•K for aluminum and 16.3 W/m۟۬•K for steel), are predicted. Compared the results of three cases, a higher thermal conductivity can also contribute more greatly to heat transfer enhancement. In T-junctions filled with porous metal materials, thermal conductivity is weakened considerably by the turbulence mixing of hot and cold fluid. The temperature fluctuation are no obvious different.

Abstract: Velocity fluctuations in a mixing T-junction were simulated in FLUENT using large-eddy simulation (LES) turbulent flow model with sub-grid scale (SGS) Smagorinsky–Lilly (SL) model. The normalized mean and root mean square velocities are used to describe the time-averaged velocities and the velocities fluctuation intensities. Comparison of the numerical results with experimental data shows that the LES model is valid for predicting the flow of mixing in a T-junction junction. The numerical results reveal the velocity distributions and fluctuations are basically symmetrical and the fluctuation at the upstream of the downstream of the main duct is stronger than that at the downstream of the downstream of the main duct.

Abstract: In the present work the temperature fluctuations in a mixing tee were simulated on FLUENT platform using the large-eddy simulation (LES) turbulent flow model with three kinds of sub-grid scale (SGS) models such as Smagorinsky-Lilly (SL) model, Wall-adapted Local Eddy-viscosity (WALE) model, and Kinetic-energy transport (KET) model. The normalized mean and root mean square temperatures were predicted and analyzed with consideration of buoyancy. The numerical results showed that buoyancy greatly influences the mixing flow and the thermal striping phenomena were quite obvious. These three SGS models have somewhat similar accuracies for prediction of the temperature fluctuation and thermal stripping in a tee of mixing hot and cold fluids.

Abstract: In the present paper, large-eddy simulation (LES) based on commercial computational fluid dynamics (CFD) software FLUENT for prediction of flow and heat transfer in a mixing T-junction was completed. Mean and root mean square (RMS) temperature and velocity were defined to describe the distributions and fluctuations of temperature and velocity. Numerical results indicate that profiles between symmetrical planes are almost same and the root mean square temperature and velocity close to the center of the main duct in the downstream are larger than those near the main duct wall. The prediction of the fluctuations of temperature and velocity is significant to understand the knowledge of the cause of thermal fatigue in a mixing T-junction.

Abstract: The velocity fields of turbulent flow in Tee junctions have been calculated using large-eddy simulations model for three cases of various inlet velocities with periodic porous media. The ratio of the inlet velocity of the main tube to that of the branch tube were respectively 0.5, 1 and 2 in three cases. In the porous medium region of the Tee junction, the velocity at the center is obvious larger than that at the top or the bottom and the pressure drop heightens rapidly. Comparison of the three cases, the pressure drop, mean velocity, normalized velocity fluctuation and the velocity oscillation versus time are increase as the inlet velocity of the main tube increase.

Abstract: On the basis of superstructure of heat exchanger network (HEN), we established a particle swarm optimization (PSO) model of HEN with no splits, with the target of minimizing investment and operation cost. A typical HEN was solved via a modified particle swarm optimization (PSO). Through comparative of the optimization result, we could know that this method could reach better solution accuracy.

Abstract: Three-dimensional models of heat transfer have been established and numerically solved using a commercial software package, Fluent, in order to obtain distributions of temperature, velocity, pressure, and liquid volume fraction of the polymer. The influences of the boundary conditions on the phase change of the polymer and the temperature distribution in the die have been evaluated. The results show that the temperature of the region close to the pelletizing surface is relatively low due to the cooling effect of the cool water, while the temperature deeper inside the die is higher, with a lower temperature gradient, as a result of the heating effect of the hot thermal oil and the polymer. A solidification phase change of the polymer occurs near the polymer outlet due to heat loss from the polymer to the water, while deeper inside the hole the polymer remains fluid without solidification, due to heating by the thermal oil. Numerical simulation provides a reliable method to optimize the design of the die, the choice of metallic material for the die, and the operating conditions of the polymer pelletizing under water.