Search

Analysis of the slipstream development around the high-speed trains in tunnels would provide references for assessing the transient gust loads on trackside workers and trackside furniture in tunnels. This paper focuses on the computational analysis of the slipstream caused by high-speed trains passing through double-track tunnels with a cross-sectional area of 100 m2. Three-dimensional unsteady compressible Reynolds-averaged Navier-Stokes equations and a realizable k-ε turbulence model were used to describe the airflow characteristics around a high-speed train in the tunnel. The moving boundary problem was treated using the sliding mesh technology. Three cases were simulated in this paper, including two tunnel lengths and two different configurations of the train. The train speed in these three cases was 250 km/h. The accuracy of the numerical method was validated by the experimental data from full-scale tests, and reasonable consistency was obtained. The results show that the flow field around the high-speed trains can be divided into three distinct regions: the region in front of the train nose, the annular region and the wake region. The slipstream development along the two sides of train is not in balance and offsets to the narrow side in the double-track tunnels. Due to the piston effect, the slipstream has a larger peak value in the tunnel than in open air. The tunnel length, train length and length ratio affect the slipstream velocities; in particular, the velocities increase with longer trains. Moreover, the propagation of pressure waves also induces the slipstream fluctuations: substantial velocity fluctuations mainly occur in front of the train, and weaken with the decrease in amplitude of the pressure wave. [ABSTRACT FROM AUTHOR]

Two-dimensional numerical simulations are carried out to examine the problem of transient electroconvection stability of dielectric liquids subjected to unipolar injection. The entire set of electrohydrodynamics equations associated with the electroconvective phenomena that occur in a layer of a dielectric liquid between two parallel electrodes subjected to a potential difference are solved numerically. We first validate the numerical simulation by comparing our linear stability electroconvection criteria with those obtained by other authors with a stability approach. In this paper, we restrict the study to the strong injection case, which corresponds to values of the non-dimensional injection parameter C greater than or equal to 10. The numerical solution of the electroconvective problem is then presented for rigid lateral boundary conditions. A detailed analysis of the scenario that occurs for different characteristic values of the stability parameter T is provided. The flow structure and its behaviour highlight the existence of different regimes, from laminar to chaotic. The development of charged plumes has been observed in particular. We compute the electrical Nusselt number for different values of the stability parameter and ion mobility. The electrical Nusselt number saturates with increasing T, a fact that it is in agreement with available experimental data. Finally, a spectral analysis is conducted for different aspect ratios of the computational domain. The spectral analysis gives an insight into the physical origin of the velocity and current oscillations. [ABSTRACT FROM AUTHOR]

This paper presents a review of computational uncertainties in scientific computing, as well as quantification of these uncertainties in the context of numerical simulations for thermo-fluid problems. The need for defining a measure of the numerical error that takes into account errors arising from different numerical building blocks of the simulation methods is discussed. In the above context, the effects of grid resolution, initial and boundary conditions, numerical discretization, and physical modeling constraints are presented. [ABSTRACT FROM AUTHOR]

The stretched-vortex subgrid model is used to run large-eddy simulations of temporal mixing layers at various Reynolds and Schmidt numbers, with different initial and boundary conditions. A self-similar flow is obtained, during which the growth rate, mean velocity and Reynolds stresses are in accord with experimental results. However, predictions of the amount of mixed fluid, and of the variation in its composition across the layer, are excessive, especially at high Schmidt number. More favourable comparisons between experiment and simulation are obtained when the large-scale flow is quasi-two-dimensional; however, such states are not self-similar and not sustainable. Present model assumptions lead to predictions of the continued subgrid spectrum with a viscous cutoff that is dependent on grid resolution. [ABSTRACT FROM AUTHOR]

A stochastic Lagrangian approach is proposed to describe dispersion in a two-dimensional low-regime river flow past random bed undulations characterized by the superposition of periodical and stationary exponential correlations, through a suitable time-dependent coefficient. The resulting dimensionless expression depends on the overall resistance factor, which is proportional to the average Peclet number of the process. A graphical analysis shows that the oscillatory transient originating from the periodic component of the bottom elevation pattern is enhanced by reduced global flow resistance, while relatively more intense tracer spreading is associated with wavelike profiles affected by persistent trendless random noise, which also determines the characteristic time needed by the plume to achieve the asymptotic domain. Numerical simulations, validating the first-order analytical approach in a wide range of heterogeneous bed geometries, are also discussed. A semianalytical procedure is finally suggested for the study of depth-averaged transport processes in real three-dimensional streams, based on the use of the derived dispersion coefficient. [ABSTRACT FROM AUTHOR]