Search results

The purpose of this paper is to develop a three‐dimensional (3D) numerical model capable of predicting the vaporization rate of a liquid fuel droplet exposed to a convective turbulent airflow at ambient room temperature and atmospheric pressure conditions.

The 3D Reynolds‐Averaged Navier‐Stokes equations, together with the mass, species, and energy conservation equations were solved in Cartesian coordinates. Closure for the turbulence stress terms for turbulent flow was accomplished by testing two different turbulence closure models; the low‐Reynolds number (LRN) k‐ε and shear‐stress transport (SST). Numerical solution of the resulted set of equations was achieved by using blocked‐off technique with finite volume method.

The present predictions showed good agreement with published turbulent experimental data when using the SST turbulence closure model. However, the LRN k‐ε model produced poor predictions. In addition, the simple numerical approach employed in the present code demonstrated its worth.

The present study is limited to ambient room temperature and atmospheric pressure conditions. However, in most practical spray flow applications droplets evaporate under ambient high‐pressure and a hot turbulent environment. Therefore, an extension of this study to evaluate the effects of pressure and temperature will make it more practical.

It is believed that the numerical code developed is of great importance to scientists and engineers working in the field of spray combustion. This paper also demonstrated for the first time that the simple blocked‐off technique can be successfully used for treating a droplet in the flow calculation domain.

This paper aims to describe the numerical simulation of a three‐dimensional turbulent free jet issuing from a sharp‐edged equilateral triangular orifice into still air.

The numerical simulation was carried out by solving the governing three‐dimensional Reynolds‐averaged Navier‐Stokes equations. Several two‐equation eddy‐viscosity models (i.e. the standard k‐ε, renormalization group (RNG) k‐ε, realizable k‐ε, shear‐stress transport (SST) k‐ω), as well as the Reynolds stress models (i.e. the standard RSM and the SSG) were tested to simulate the flowfield. The numerical predictions were compared with experimental data in order to assess the capability and limitations of the various turbulent models examined in this work. Findings –The vena contracta effect was predicted by all the tested models. Among the eddy‐viscosity models only the realizable k‐ε model showed good agreement of the near‐field jet decay. None of the eddy‐viscosity models was capable of predicting the profiles of the jet turbulence intensities. The RSMs, especially the standard RSM, were able to produce much better predictions of the features of the jet in comparison with the eddy‐viscosity models. The standard RSM predictions were found to agree reasonably well with the experimental data.

The conclusion, that among the tested RANS turbulence closure models, the RSM appeared the only one capable of reproducing reasonably well the experimental data concerns only the jet flow case examined here. Also, the average computational time for a single run was quite long, i.e. 340 h, but it is believed that parallel computing will reduce it considerably.

The numerical results reported in this paper provide a comparison between several RANS turbulence closure models for simulating a turbulent free jet issuing from an equilateral triangular nozzle.

This paper aims to introduce a two-dimensional smoothed particle hydrodynamics (SPH) framework for simulating the evaporation and combustion process of fuel droplets.

To solve the gas–liquid two-phase flow problem, a multiphase SPH method capable of handling high density-ratio problems is established. Based on the Fourier heat conduction equation and Fick’s law of diffusion, the SPH discrete equations are derived. To effectively characterize the phase transition problem, inspired by volume of fluid method, the concept of liquid phase mass fraction of the SPH particles is proposed. The one-step global reaction model of n-hexane is used for the vapor combustion.

The evaporation and combustion process of single droplet conforms to the law. The framework works out well when the evaporation of multiple droplets involves coalescence process. Three different kinds of flames are observed in succession in the combustion process of a single droplet at different inflow velocity, which agree well with the results of the experiment.

To the best of the authors’ knowledge, this is the first computational framework that has the capability to simulate evaporation and combustion with SPH method. Based on the particle nature of SPH method, the framework has natural advantages in interface tracking.