Hybrid Flame Combustion Burner

The diffusion and Premixed combustion types classified according to the provide way of air and fuel inside the burner, have collision advantages in terms of emissions of pollutant and stability of flame. The advantage of diffusion flame is higher stability, while the second type (premixed flame) is minimum emissions its means clean combustion as compared with another types. Due to the flame stability in practical combustor represents significant industrial design factor. Recently, to meet the increasingly emission controls, The recently toward to the premixed combustion so its application has been gradually enhancement with numerous advantages like shorter flame length burner, higher thermal efficiency, likewise low emissions level as compared with the type of diffusion flame [1]. Even so, the stability of flame (flash-back and lifting-limits) is limited in the premixed combustion type. There are many strategies to enhance the stability of the premixed flame by creating swirl flow, cyclone configuration and pilot flames. The cyclone configuration grants a flow in a swirl shape due to air/fuel ratio tangential at inlet consequently develops the mixing of air-fuel and the stability level of flame by increasing chemical species recirculation [2, 3]. Otherwise, the combustor of cyclone configuration has a simpler burner design and the intensity of swirl is high as compared to a traditional swirl burner (complicated swirl design) [4, 5]. Ishizuka et al. [6] and Shimokuri et al. [7] investigated the premixed flame type in tube experimentally which was created by using tangentially slots through the tube.

They discovered thermally stable of tubular flame due to the structure symmetrically of chemical surroundings, and the aerodynamically stable due to suitable heat relief and pressure distributions for break the instabilities of combustion. Even so, slit width or slit length for tubular combustion type are additional improvements to complete gases burn at high velocities (50 meter/sec) because the little combustion residence time. Onuma et al. [8] improved a cyclone configuration burner to meet the high level of stability of premixed& diffusion flame and minimum pollution. The fuel of premixed flame provided into tangentially slots of the burner to utilize that technique to holder of flame which its work to stabilize the type of diffusion flame. The emission could be closely degreased of non-diffusion combustion by enhance the mixing rate. The hybrid flame combustion type using the premixed and diffusion combustion types together, its supply the two benefits in terms of pollutant emissions and stability of flame. Hwang et al. [9] and Noda et al. [10] achieved experimentally the high level of flame stability and low pollutant emissions at the cyclone hybrid burner. By means of developing the configuration design of the fuel injection for the diffusion flame, it noted that an enhancement in mixing rate of reactions by using a multi slots fuel injection to grants a stability of flame approximately double as wide as that of the fuel injection when using a single slot. Also, an optimized burner investigating by hybrid flame combustion type yielded NOx combustor for clean burning and effective [11].

3.1 Mathematical Concept

A modeling of mathematical turbulent flow is achieved by using partial differential equations (PDE) to the derivative of nonlinear equation which represent mass, chemical species, and momentum conservation and energy equations for 2D burner in surrounding [12].

The conservation of mass equation in form of:

$\frac{\partial \rho}{\partial t}+\frac{\partial}{\partial x_j}\left(\rho V_j\right)=0$   (1)

where, ρ: Density, V_j: The instantaneous velocity at different direction of j, where (j = 1, 2, 3).

The averaged momentum equation is given as follows:

$\begin{aligned} & \frac{\partial}{\partial t}\left(\rho V_j\right)+\frac{\partial}{\partial x_j}\left(\rho V_i V_{\mathrm{j}}\right)=-\frac{\partial \mathrm{p}}{\partial \mathrm{x}_{\mathrm{j}}}+\frac{\partial \tau_{\mathrm{ij}}}{\partial \mathrm{x}_{\mathrm{j}}}+\frac{\partial}{\partial \mathrm{x}_{\mathrm{j}}}\left(-\rho \bar {\grave{V}_{\mathrm{i}} \grave{V}_{\mathrm{j}}}\right) +F_j & \end{aligned}$   (2)

where, τ_i: Stress tensor which written as follow:

$\tau_{i j}=\mu\left[\left(\frac{\partial \mathrm{V}_{\mathrm{i}}}{\partial \mathrm{x}_{\mathrm{j}}}+\frac{\partial \mathrm{V}_{\mathrm{j}}}{\partial \mathrm{x}_{\mathrm{j}}}\right)-\frac{2}{3} \delta_{\mathrm{ij}}\left(\frac{\partial \mathrm{V}_{\mathrm{k}}}{\partial \mathrm{x}_{\mathrm{k}}}\right)\right]$   (3)

And, -ρV_i^’ V_j^’: Reynolds stresses which expressed by: where, P: Pressure, µ: Fluid dynamic viscosity, µ_t: Turbulent viscosity.

The energy equation is given by:

$-\rho \overline{\grave{V}_{\mathrm{i}} \grave{V}_{\mathrm{j}}}=\mu_{\mathrm{t}}\left(\frac{\partial \mathrm{V}_{\mathrm{i}}}{\partial \mathrm{x}_{\mathrm{j}}^{\mathrm{}}}+\frac{\partial V_{\mathrm{j}}}{\partial \mathrm{x}_{\mathrm{j}}}\right)-\frac{2}{3} \delta_{\mathrm{ij}}\left(\rho_{\mathrm{k}}+\mu_{\mathrm{t}} \frac{\partial \mathrm{V}_{\mathrm{j}}}{\partial \mathrm{x}_{\mathrm{j}}}\right)$   (4)

$\frac{\partial(\rho h)}{\partial t}+\frac{\partial}{\partial \mathrm{x}_{\mathrm{j}}}\left(\rho \mathrm{V}_{\mathrm{j}} \mathrm{h}\right)=\frac{\partial}{\partial \mathrm{x}_{\mathrm{j}}}\left(\rho \mathrm{D}_{\mathrm{h}}\right) \frac{\partial \mathrm{h}}{\partial \mathrm{x}_{\mathrm{j}}}-\rho \overline{{\grave{h}} \grave{V}_{\mathrm{j}}}$    (5)

where, D_h=λ/ρc_p: Thermal diffusivity, -ρh^’ V_j^’: Correlation term between the enthalpy(h) and fluctuation of velocity.

Mass enthalpy h_k for each species written as follow:

$\mathrm{h}=\sum_{\mathrm{k}=1}^{\mathrm{n}} \mathrm{Y}_{\mathrm{k}} \mathrm{h}_{\mathrm{k}}$    (6)

Chemical species conservation equation for reactive mixture is given by:

$\frac{\partial\left(\rho Y_k\right)}{\partial t}+\frac{\partial\left(\rho V_{\mathrm{i}} Y_k\right)}{\partial {x}_j}=\frac{\partial J_{\mathrm{j}}^{\mathrm{k}}}{\partial {x}_{\mathrm{j}}}+\rho \grave{\mathrm{w}}_{\mathrm{k}}$   (7)

where, Y_k, w_k, J_j^k are mass fraction, k: Species of the production rate & mass diffusion flux in direction of j. The chemical equation of combustion liquid petroleum gas LPG (40% propane, 60% butane) has been given by:

$\begin{gathered}0.4 \mathbf{C}_3 \mathbf{H}_8+0.6 \mathbf{C}_4 \mathbf{H}_{10}+5.9\left(\mathbf{O}_2+3.76 \mathbf{N}_2\right) \\ 3.6 \mathbf{C O}_2+4.6 \mathbf{H}_2 \mathbf{O}+22.184 \mathbf{N}_{\mathbf{2}}\end{gathered}$    (8)

3.2 Modeling of turbulence

In turbulent the Reynolds revealed new unknown term which it’s part of Navier-Stokes equations (the non-linearity). -ρV_i^’ V_j^’ represents correlation between directions i & j and fluctuations speeds, -ρh^’ V_j^’ represents correlation between the mass enthalp and fluctuations speeds. The standard k-ε model is interested in this study.

The standard k-ε turbulence model: The k-ε standard turbulence model is consist of 2 equations, the first one is the kinetic energy of turbulent and the second one grants the characteristic of scale length of the turbulent. The kinetic energy of turbulent k & the kinetic energy of turbulent of the dissipation rate ε are given as follows:

$\frac{\partial(\rho \mathrm{k})}{\partial \mathrm{t}}+\frac{\partial\left(\rho \mathrm{V}_{\mathrm{i}} \mathrm{k}\right)}{\partial \mathrm{x}_{\mathrm{j}}}=\frac{\partial}{\partial \mathrm{x}_{\mathrm{j}}}\left[\left(\mu+\frac{\mu_{\mathrm{t}}}{\sigma_{\mathrm{k}}}\right) \frac{\partial \mathrm{k}}{\partial \mathrm{x}_{\mathrm{j}}}\right]+\mathrm{P}_{\mathrm{k}}-\rho \varepsilon$    (9)

$\begin{gathered}\frac{\partial(\rho \varepsilon)}{\partial \mathrm{t}}+\frac{\partial\left(\rho \mathrm{V}_{\mathrm{j}} \varepsilon\right)}{\partial \mathrm{x}_{\mathrm{j}}}=\frac{\partial}{\partial \mathrm{x}_{\mathrm{j}}}\left[\left(\mu+\frac{\mu_{\mathrm{t}}}{\sigma_{\mathrm{k}}}\right) \frac{\partial \varepsilon}{\partial \mathrm{x}_{\mathrm{j}}}\right] \\ +\mathrm{C}_{\varepsilon 1} \frac{\varepsilon}{\mathrm{k}} \mathrm{P}_{\mathrm{k}}-\mathrm{C}_{\varepsilon 2} \rho \frac{\varepsilon^2}{\mathrm{k}}\end{gathered}$    (10)

where, the term of turbulent viscosity & the term of turbulent kinetic energy are defined by:

$\mathrm{P}_{\varepsilon}=\mathrm{C}_{\varepsilon 1} \frac{\varepsilon}{\mathrm{k}} \mathrm{P}_{\mathrm{k}}$    (11)

$\mu_{\mathrm{t}}=\rho \mathrm{C}_\mu \frac{\mathrm{k}^2}{\varepsilon}$    (12)

3.3 Combustion modeling

The turbulent no-premixed flame is interested in this study. Thus, the model of Eddy Dissipation was chosen to seeking the rate of reactions and observation the turbulence interaction. This model has showed for turbulent diffusion flames at fast reactions infinitely. The origin expression of production of species i because the reaction is presented the lower of mixing rate of product & mixing rate of the reactant:

$\mathrm{S}_{\mathrm{mf}}=-\rho \frac{\varepsilon}{\mathrm{k}} \min \left\lceil\grave{\mathrm{u}}_{\mathrm{i}} \mathrm{M}_{\mathrm{w}, \mathrm{i}} \mathrm{A}\left(\frac{\mathrm{Y}_{\mathrm{R}}}{\grave{u}_{\mathrm{R}} \mathrm{M}_{\mathrm{w}, \mathrm{R}}}\right), \grave{\mathrm{u}}_{\mathrm{i}} \mathrm{M}_{\mathrm{w}, \mathrm{i}} \mathrm{AB} \frac{\sum_{\mathrm{P}} \mathrm{Y}_{\mathrm{P}}}{\sum_{\mathrm{j}}^{\mathrm{N}} \grave{\mathrm{u}}_{\mathrm{j}} \mathrm{M}_{\mathrm{w}, \mathrm{j}}}\right\rceil$    (13)

3.4 Procedure of discretization

The commercial CFD code ANSYS Fluent 16.1 used in this study, model the combustion in no-premixed flame [(40%propane+60%butane)-air] with a cyclone burner. The governing equations of numerical solution which including mass equation, momentum equation, energy equation, and species of (reaction and products) investigated by a finite volume method. The pressure and velocity coupling algorithm as a simple formula as well as the upwind scheme of discretization are applied to resolve the governing equations. The Eddy dissipation model applied to consider the reactions turbulent.

3.5 Boundary conditions and meshing

The experimental test rig was implemented and studied premixed, diffusion and hybrid flame burner. The comparison of three types of flame in experimental work was investigated. The diffusion flame burner was numerically achieved by cyclone burner which its configuration is axis-symmetric. The of cylindrical coordinates of (computational fluid dynamics) CFD Ansys Fluent 16.1. This paper is concerned to make comparison the numerical results of diffusion flame with experimental result. A rough grid of numerical results had a big different from the experimental datum. Actually, the mathematical results are closing to the experimental works with the grid gradual refining. Thus, the using of no-uniform grid spacing grants a precise resolution near to the central of the cyclone burner. For the full domain simulation, a control volume is divided into uninform quadrilateral in order to resolve the steep gradients and to conserve flux in each cell. A total number of 34,500 cells with 3,590 nodes were used for the simple cylindrical injector as shown in Table 2 of grid parameter for half symmetric burner. The design of the geometry of the volume control and the mesh generation was performed using the software "GAMBIT" with a quadratic mesh. A refined mesh in the area near the exit of the burner is represented to fully model the differences phenomena. The other order which inverses scheme used the convective terms of momentum equations, the turbulent kinetic energy and the dissipation rate of the turbulent kinetic energy. For speed correction pressure the simple algorithm is used.

From experimental test rig configuration, the boundary conditions are determined as shown in Figure 5. The inlet of fuel is represented by I (velocity inlet). The air inlet represented region II (pressure inlet). The outlet of flame is represented region III to the atmospheric (pressure outlet), IV burner wall. Air temperature and gas temperature are equal (T=300 K) with (p=101325 Pa) the atmospheric pressure.

Table 2. Grid parameter

5fig.png

Figure 5. Geometry mish

Combustion of a cyclone hybrid combustor using a complex of premixed (swirling) and diffusion flames (jet) were experimentally achieved high stability of flame and low pollutant emissions. The main conclusions this study:

a. The comparisons in shape of flame between the hybrid and diffusion flame type which its represent indicate of blow-out regions.

b. The flame temperature at hybrid combustion is greater than in diffusion and premixed combustion, also the flame temperature which produced from nozzle (1) is greater than flame temperature at nozzle (2) for both hybrid and diffusion type.

c. Due to no flash-back and blow-out phenomena the hybrid flame limit of stability is large, and high-level stability at flame boarder.

d. A high swirl number was investigated by using premixed mode by tangential nozzles.

e. High temperature of diffusion flame which investigated numerically as compared with experimentally due to no heat transfer through burner.