CN106339551A - Simulation method and device for generation value of combustion characteristic of pulverized coal of power station boiler - Google Patents

Abstract

The invention discloses a simulation method and a simulation device for a generation value of a combustion characteristic of pulverized coal of a power station boiler, and solves the technical problem that the whole rule characteristics of combustion, flowing and heat transfer in the boiler cannot be measured due to the fact that a conventional hearth cannot be used for manufacturing a test bed of a full-size model as the size of the hearth is larger and larger. The method comprises the following steps: simulating a combustion process of the pulverized coal of the boiler in a preset boiler combustor model subjected to gridding division through a basic conservation equation, a turbulent flow model, a turbulent gas-solid two phase flow model, a gas phase turbulent combustion model, a pulverized coal particle combustion model, a radiation heat exchange model and an NOX generation model; determining a gas flow field in a boiler combustor, a temperature field in the boiler combustor, flue gas component distribution in the boiler combustor, a moving trace of pulverized coal particles in the boiler combustor, radiation heat flow on the wall surface of the boiler combustor, the boiler combustion efficiency and an NOX discharge amount according to a simulation result.

Description

A kind of power boiler breeze combustion characteristics generates analogy method and the device of numerical value

Technical field

The present invention relates to technical field of electric power, more particularly, to a kind of simulation of power boiler breeze combustion characteristics generation numerical value Method and device.

Background technology

Power boiler breeze burning is the three dimensional process of a complicated turbulent flow, heat transfer and burning.Due to process Complexity and the polytropy of coal-fired boiler fuel, so far, the design to boiler and operation lack ripe theory and warp Test, generally require coldandtest come to determine operation and design parameter because test have the advantages that directly perceived, reliable, can To be used directly to instruct the design of boiler product to produce and to manufacture.But current Power Plant in China is to high parameter, great Rong Amount aspect develops, and size of burner hearth is increasing, and this kind of test period is long, costly, and hardly results in comprehensive, satisfied number According to.The testing stand manufacturing full scale model is unrealistic, the heavy duty boiler of live actual motion is directly carried out with air and moves The measurement in the field of force, and burning, flowing, heat transfer global regularity characteristic in stove are measured and be nearly impossible, so, lead to Overtesting instructs boiler design to there is significant limitation.

Therefore, how above-mentioned refer to due to size of burner hearth increasing, the testing stand manufacturing full scale model is not cut Reality, the technical problem that burning, flowing, heat transfer global regularity characteristic in stove cannot be measured being led to has become this Skilled person's technical problem urgently to be resolved hurrily.

Content of the invention

A kind of power boiler breeze combustion characteristics provided in an embodiment of the present invention generates analogy method and the device of numerical value, its In, the analogy method that power boiler breeze combustion characteristics generates numerical value includes: carries out partition network to preset boiler-burner model Lattice divide；By basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas phase turbulance combustion model, coal Powder particles combustion model, radiative heat transfer model and nox generation model are in the preset boiler-burner model after stress and strain model The combustion process of boiler coal-ash is simulated；Airflow field in boiler-burner, boiler combustion are determined according to the result after simulation Smoke components distribution in temperature field in device, boiler-burner, movement locus in boiler-burner for the pulverized coal particle, boiler Burner wall surface radiation hot-fluid, boiler combustion efficiency and no_xDischarge capacity.In the present embodiment, by preset boiler-burner mould Type carries out multiblock technique division；Rapid by basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas phase Stream combustion model, pulverized coal particle combustion model, radiative heat transfer model and nox generation model are to the preset boiler combustion after stress and strain model The combustion process of the boiler coal-ash in burner model is simulated；Air-flow in boiler-burner is determined according to the result after simulation Temperature field in field, boiler-burner, the smoke components distribution in boiler-burner, fortune in boiler-burner for the pulverized coal particle Dynamic rail mark, boiler-burner Wall Radiation hot-fluid, boiler combustion efficiency and no_xDischarge capacity, solves at present due to size of burner hearth Increasing, the testing stand manufacturing full scale model is unrealistic, and led to cannot be whole to burning, flowing, heat transfer in stove The technical problem that body rule characteristic measures.

Brief description

In order to be illustrated more clearly that the embodiment of the present invention or technical scheme of the prior art, below will be to embodiment or existing Have technology description in required use accompanying drawing be briefly described it should be apparent that, drawings in the following description be only this Some embodiments of invention, for those of ordinary skill in the art, without having to pay creative labor, also may be used So that other accompanying drawings are obtained according to these accompanying drawings.

Fig. 1 is the one of the analogy method that a kind of power boiler breeze combustion characteristics provided in an embodiment of the present invention generates numerical value The schematic flow sheet of individual embodiment；

Fig. 2 is the one of the analog that a kind of power boiler breeze combustion characteristics provided in an embodiment of the present invention generates numerical value The structural representation of individual embodiment；

Fig. 3 (a) to (c) is chamber structure and stress and strain model schematic diagram；

Fig. 4 is boiler-burner y=9.56m section speed vector figure；

Fig. 5 is the speed field pattern in middle level burner and ofa region；

Fig. 6 is away from front-back wall burner outlet 0.1m speed vector figure；

Fig. 7 be middle level burner and ofa region temperature cloud picture；

Fig. 8 is y=9.56m section temperature cloud atlas；

Fig. 9 is away from burner outlet 0.3m temperature cloud picture；

The concentration distribution cloud atlas of smoke components on the vertical section that Figure 10 is located for y=9.56m burner；

Figure 11 (a) to (c) is o₂、co、no_xThe concentration of smoke components on the section that middle level burner and burnout degree are located Cloud charts；

Figure 12 is pulverized coal particle movement locus figure；

Figure 13 is the radiant heat flux schematic diagram of burner You Qiang, Hou Qiang and left wall.

Specific embodiment

Embodiments provide analogy method and the device that a kind of power boiler breeze combustion characteristics generates numerical value, solution Determine at present because size of burner hearth is increasing, the testing stand manufacturing full scale model is unrealistic, and led to cannot be right The technical problem that in stove, burning, flowing, heat transfer global regularity characteristic measure.

Cfd, software (computational fluid dynamics), i.e. computational fluid dynamicses, are hydromechanical One branch, abbreviation cfd.Cfd is modern age hydrodynamics, the product of numerical mathematics and computer science combination, is one and has The boundary science of powerful vitality.It applies the mathematical method of various discretizations, Fluid Mechanics with electronic computer as instrument All kinds of problems carry out numerical experiment, computer mould fits analysis and research, to solve various practical problems.

For enabling the goal of the invention of the present invention, feature, advantage more obvious and understandable, below in conjunction with the present invention Accompanying drawing in embodiment, is clearly and completely described it is clear that disclosed below to the technical scheme in the embodiment of the present invention Embodiment be only a part of embodiment of the present invention, and not all embodiment.Based on the embodiment in the present invention, this area All other embodiment that those of ordinary skill is obtained under the premise of not making creative work, broadly falls into present invention protection Scope.

Refer to Fig. 1, a kind of power boiler breeze combustion characteristics provided in an embodiment of the present invention generates the simulation side of numerical value One embodiment of method includes:

101st, multiblock technique division is carried out to preset boiler-burner model；

With the development of the subject such as computer technology and Fluid Mechanics Computation, numerical heat transfer, calculating Combustion, calculate Machine analogue technique is developed rapidly.Numerical simulation based on cfd is increasingly becoming the research of various countries' field of energy source power Persons are used for studying the important means of boiler boiler flow field.Method for numerical simulation speed is fast, containing much information of acquisition, can be complete and pre- Flowing in report stove, heat transfer and combustion process, provide important reference frame for the design of boiler, operation and transformation, have weight The engineering application value wanted.Thus, calculated by the full simulation value of boiler flow field, the aerodynamic field in analytical furnace, temperature , the research meanses becoming feasible that the generation of nox in combustion process is given a forecast are it is necessary first to preset boiler-burner Model carries out multiblock technique division.

102nd, basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas phase turbulance burning mould are passed through Type, pulverized coal particle combustion model, radiative heat transfer model and nox generation model are to the preset boiler-burner model after stress and strain model In boiler coal-ash combustion process be simulated；

After multiblock technique division is carried out to preset boiler-burner model, need by basal conservation eqution, turbulent flow Flow model, gas-particle two-phase flow movable model, gas phase turbulance combustion model, pulverized coal particle combustion model, radiative heat transfer model With nox generation model, the combustion process of the boiler coal-ash in the preset boiler-burner model after stress and strain model is simulated.

103rd, the airflow field in boiler-burner, the temperature field in boiler-burner, pot are determined according to the result after simulation Smoke components distribution in burner, movement locus in boiler-burner for the pulverized coal particle, boiler-burner Wall Radiation Hot-fluid, boiler combustion efficiency and no_xDischarge capacity.

When by basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas phase turbulance burning mould Type, pulverized coal particle combustion model, radiative heat transfer model and nox generation model are to the preset boiler-burner model after stress and strain model In boiler coal-ash combustion process be simulated after, need according to simulation after result determine the air-flow in boiler-burner Temperature field in field, boiler-burner, the smoke components distribution in boiler-burner, fortune in boiler-burner for the pulverized coal particle Dynamic rail mark, boiler-burner Wall Radiation hot-fluid, boiler combustion efficiency and no_xDischarge capacity.

It should be noted that also including after multiblock technique division is carried out to preset boiler-burner model:

Preset boiler-burner model after stress and strain model is simplified, and gets the design parameter after setting and work Condition condition.

Alternatively, turbulent flow model is realizable k- ε two-equation model；

Wherein, k equation is

ε equation is

Gk represents that the Turbulent Kinetic being caused due to average velocity gradient is produced, and gb is to move for the turbulent flow that buoyancy effect causes Can produce, ym represents the impact that fast turbulence pulsation can be pressed to expand to total dissipative shock wave, σ_k、σ_εIt is Turbulent Kinetic and its dissipation respectively The turbulent prandtl number of rate.

Alternatively, gas-particle two-phase flow movable model is Lagrangian stochastic particle model trajectory.

Alternatively, gas phase turbulance combustion model is Hybrid analysis probability density function (pdf) model.

Alternatively, pulverized coal particle combustion model include for fugitive constituent separate out two-equation model, for coke burning Diffusion kinetics control combustion model.

Alternatively, radiative heat transfer model is p-1 radiation patterns

Alternatively, nox generation model is pdf transport equation model.

To be described with a specific embodiment below, application examples includes:

For power station large coal-fired boiler combustion process, basal conservation eqution, turbulent flow model, turbulent flow gas-solid can be used Two-phase flow model, gas phase turbulance combustion model, pulverized coal particle combustion model, radiative heat transfer model, to carry out mathematical description, lead to Cross to simulate real process to above-mentioned equation numerical solution, provide reference parameter for actual motion.This part is just to this report mould Each mathematical model intending the coal powder boiler combustion process employed in research describes in detail.

Basal conservation eqution

Burning is the turbulent flow process including very exothermic chemical reaction, and all physical quantitys are all room and times Stochastic variable, but turbulent flow follows continuous media general motion rule.The law of description combustion law has: the conservation of mass, The conservation of momentum, preservation of energy, chemical constituent balance and chemical element conservation of mass etc..

(1) equation of continuity

$\frac{\∂\mathrm{\ρ}}{\∂t}+\frac{\∂}{\∂x_j}\left({\mathrm{\ρu}}_j\right)=0---(2-1)$

(2) equation of momentum

The general type of the equation of momentum can be written as:

σ in formula_ij=p δ_ij-τ_ij

Wherein:τ_ijFor viscous stress.

δ_ijFor Kronecker function:

δ_ijFor stress tensor, s_iThen include various body forces and the component in i direction for the resistance.Considering multiphase flow When, the active force between multiphase flow also be reflected in this.

(3) energy equation

Energy-balance equation:

The equation equal sign left side: represent the rate of change to the time for the unit fluid gross energy in the unit interval；Equation equal sign The right: the 1st is surface pressing convection cell infinitesimal work done, typically negligible；2nd unit volume causing for conduction of heat Energy variation；3rd φ is because viscous effect mechanical energy is converted into the part of heat energy, referred to as dissipative function；4th s_hFor changing Learn reaction heat, radiant heat, alternate heat exchange and self-defining volumetric sources item.

(4) chemical multi-species equation

What chemical multi-species equation embodied is the conservation of mass of each component in combustion process.For any chemical constituent K, its chemical constituent equation of continuity is:

$\frac{\∂}{\∂t}\left({\mathrm{\ρm}}_k\right)+\frac{\∂}{\∂x_j}\left({\mathrm{\ρu}}_j{m}_k\right)=\frac{\∂}{\∂x_j}\left({\mathrm{\γ}}_k\frac{\∂m_k}{\∂x_j}\right)+s_k---(2-4)$

In formula: m_kThe mass fraction of component k, is defined as:

s_kThis component that generation (or consumption) rate of component k being caused due to chemical reaction and heterogeneous reaction are produced Mass Sources.

γ_kThe transport coefficient of chemical constituent k: γ_k=ρ d_k

d_kDiffusion coefficient for the corresponding mixed gas of chemical constituent k.

Formula (2-4) is added to whole component k, that is, obtains the equation of continuity of whole fluid:

$\frac{\∂\mathrm{\ρ}}{\∂t}+\frac{\∂}{\∂x_j}\left({\mathrm{\ρu}}_j\right)=\underset{k}{\σ}{s}_k$

$\mathrm{\ρ}=\underset{k}{\σ}{\mathrm{\ρ}}_k$

In formula,The total source item of quality causing for particle reaction, when no granule phase reaction

(5) state equation

Contain 6 unknown numbers, u, v, w, p, t, and ρ in aforesaid equation, also need to supplement a contact p, the state side of ρ Journey:

ρ=ρ (p, t) (2-5)

For above-described fundamental equation, its unknown number is equal it should say that equation is closing with equation number.As long as Boundary condition and initial condition are suitably described it is possible to solve, but in fact it has been found that in actual nature and engineering stream In dynamic device, flowing is often turbulent flow, and turbulent flow is to carry out in the scale of turbulence of a very little.Therefore, solve this One group of equation of sample is necessary for carrying out in the size of mesh opening of the scale of turbulence, but this is current computer capacity and speed still can not Realize.Therefore, solve navier-stokes equation to set about in terms of other, here it is turbulence model.

Turbulent flow model

Turbulent flow is the common flow phenomenon of nature, and the flowing of fluid in most engineerings is often in turbulence-like State, turbulence characteristic occupies an important position in engineering.In boiler combustion process, because combustion apparatus size is larger, shape Complicated, air velocity is higher, adds the impact of the chemical reactions such as fuel combustion, the air current flow in burner and Boiler Furnace is several It is all turbulent flow.

K- ε model in two-equation model, is the turbulence model being most widely used in Practical Project.With regard to turbulent flow On the basis of the equation of kinetic energy k, it has been re-introduced into representing that isotropism microvortex revolves the pulsation that mechanical energy changes into heat energy speed Kinetic energy dissipation rate ε.This model is to be proposed in 1972 by launder and spalding.It is now widely used for combustor stream Field, the numerical simulation of pipe flow, bluff body backflow and jet backflow, and air current flow, burning, heat transfer, mass transfer in some equipment Deng the gas phase simulation in process synthesis model.

K- ε two-equation model has the advantage that (1) considers the transport process of turbulent flow physical quantity by seeking partial differential equation, Determine the relation of pulsatile characteristics speed and mean field velocity gradient by solving partial differential equation, rather than directly both are joined System gets up；(2) characteristic length is not by empirically determined, but using dissipative scale as characteristic length and inclined accordingly by solving The differential equation obtains.

But k- ε model is to buoyant flow, strong eddy flow moves, curved wall flows, bending streamline flows, low reynolds number turbulent flow And during the flowing such as circular jet, certain distortion can be produced, reason is in the k- ε model of standard, for reynolds stress Each component it is assumed that viscosity coefficient μ_tIt is identical, that is, suppose μ_tIt is isotropic scalar.And the situation in bending streamline Under, turbulent flow is anisotropic, μ_tIt should be anisotropic tensor.Therefore, many scholars are directed to some problems to k- ε model It is corrected, such as consider nonequilibrium flow, rotation and curvature, compressibility and non-linear etc. develop many improved moulds Type, such as low reynolds number k- ε model, non-linear k- ε model, multiple dimensioned k- ε model, renormalization group k-ε model, achievable k- ε model etc., and the model after application enhancements has been achieved for satisfied effect in some respects.

Realizablek- ε (the band rotation is revised) two-equation model that this report selects carries out numerical simulation calculation. The main feature of realizable k- ε two-equation model is exactly it is to consider turbulent flow physical quantity by seeking partial differential equation Transport process, determines the relation of pulsatile characteristics speed and mean field velocity gradient by solving partial differential equation, rather than directly Connect and both are connected；In addition, the characteristic length in model is not by empirically determined, but long as feature using dissipative scale Degree, and obtained by solving corresponding partial differential equation.Therefore the k- ε model with swirling modification is sent out for flat board and round jet Scattered ratio has more accurately to be predicted, and it is for rotational flow, the boundary layer flow of strong adverse pressure gradient, flow separation and secondary Stream has good performance.Comparatively realizablek- ε both sides journey is more simple, easy to use.Realizablek- ε now Two-equation model is own to be effectively used for various types of flow simulating, including rotation uniform shear flow, includes jet In free-flowing with mixed flow, pipeline flowing, boundary layer flow and with detached flowing etc..And all achieve with Test data than more consistent result, study by suitable engineering problem.

K- ε model with swirling modification is that just occur in the recent period, to have two main differences compared with standard k-ε model Point: (1) k- ε model with swirling modification increased a formula for turbulent viscosity, and (2) increased new transmission side for dissipative shock wave Journey, this equation derives from an Exact Equation made for laminar velocity fluctuation.By revised standard k-ε model, bright The simulation precision of the aobvious diffusibility that improve to plane jet and round jet.

Under rectangular coordinate system, the general control equation of the standard k-ε model of stable state is as follows:

$\begin{array}{c}\frac{\∂}{\∂t}\left(\mathrm{\ρ}\mathrm{\φ}\right)+\frac{\∂}{\∂x}\left(\mathrm{\ρ}u\mathrm{\φ}\right)+\frac{\∂}{\∂y}\left(\mathrm{\ρ}v\mathrm{\φ}\right)+\frac{\∂}{\∂z}\left(\mathrm{\ρ}w\mathrm{\φ}\right)\\ =\frac{\∂}{\∂x}\left(\mathrm{\γ}\frac{\∂\mathrm{\φ}}{\∂x}\right)+\frac{\∂}{\∂y}\left(\mathrm{\γ}\frac{\∂\mathrm{\φ}}{\∂y}\right)+\frac{\∂}{\∂z}\left(\mathrm{\γ}\frac{\∂\mathrm{\φ}}{\∂z}\right)+s\end{array}---(2-8)$

Wherein: p is Fluid pressure, ρ is fluid density, and φ is general dependent variable, and г is the transport coefficient of each equation variable (diffusion coefficient), s is corresponding source item in the conservation equation of dependent variable, and the particular content of p, ρ, φ, г, s is shown in Table 1.

The expression formula of each variable in table 1 general control equation

In Table 1

$g_k={\mathrm{\μ}}_t\{2\[{\left(\frac{\∂u}{\∂x}\right)}^2+{\left(\frac{\∂v}{\∂y}\right)}^2+{\left(\frac{\∂w}{\∂z}\right)}^2\]+{(\frac{\∂u}{\∂y}+\frac{\∂v}{\∂x})}^2+{(\frac{\∂u}{\∂z}+\frac{\∂w}{\∂x})}^2+{(\frac{\∂w}{\∂y}+\frac{\∂v}{\∂z})}^2\}---(2-9)$

μ_e=μ+μ_t,

Wherein: μ_eFor coefficient of virtual viscosity；

μ_tFor coefficient of eddy viscosity；

μ is laminar flow viscosity.

It should be noted that, during standard k-ε model pair, the king-sized situation of equal strain rate is it is possible to lead to negative malleation Power.For making flowing meet the physical law of turbulent flow, need to carry out certain mathematical constraint to normal pressure.Calculate turbulent kinetic limit meter C in formula_μShould not be constant, and should connect with strain rate.Thus proposing realizable k- ε model

K equation:

ε equation:

Wherein: g_kRepresent that the Turbulent Kinetic being caused due to average velocity gradient is produced, g_bIt is the rapids causing for buoyancy effect The energy of flow produces；y_mExpression can press fast turbulence pulsation to expand the impact to total dissipative shock wave.σ_k、σ_εBe respectively Turbulent Kinetic and its The turbulent prandtl number of dissipative shock wave.

c₂=1.9, σ_k=1.0, σ_ε=1.2, σ_1ε=1.44

$c_1=max(0.43,\frac{\mathrm{\η}}{\mathrm{\η}+5}),\mathrm{\η}={(2e_{ij}\·e_{ij})}^{1/2}\frac{k}{\mathrm{\ϵ}},e_{ij}=\frac{1}{2}(\frac{\∂u_i}{\∂x_j}+\frac{\∂u_j}{\∂x_i})$

In formula, μ_tWith c_μIt is calculated as follows:

${\mathrm{\μ}}_t=\frac{c_{\mathrm{\μ}}{\mathrm{\ρk}}^2}{\mathrm{\ϵ}},c_{\mathrm{\μ}}=\frac{i}{a_o+a_s{u}^{*}k/\mathrm{\ϵ}}$

Wherein:

a_o=4.0

$a_s=\sqrt{6}cos\mathrm{\φ}$

$\mathrm{\φ}=\frac{1}{3}arccos\left(\sqrt{6}w\right)$

$w=\frac{e_{ij}{e}_{jk}{e}_{ki}}{{(e_{ij}\·e_{ij})}^{1/2}}$

$u^{*}=\sqrt{e_{ij}{e}_{ij}+{\widetilde{\mathrm{\ω}}}_{ij}{\widetilde{\mathrm{\ω}}}_{ij}}$

$\begin{array}{cc}{\widetilde{\mathrm{\ω}}}_{ij}={\mathrm{\ω}}_{ij}-2{\mathrm{\ϵ}}_{ijk}{\mathrm{\ω}}_k& {\mathrm{\ω}}_{ij}={\stackrel{\&overbar;}{\mathrm{\ω}}}_{ij}-2{\mathrm{\ϵ}}_{ijk}{\mathrm{\ω}}_k\end{array}$

WhereinIt is to be ω from angular velocity_kReference frame in observe when equal slewing rate tensor it is clear that to no The flow field of rotation, u^*Section 2 in radical sign in formula is zero, and this is the impact being specifically intended for representing rotation, is also this mould One of feature of type.

Gas-particle two-phase flow movable model

Process of coal combustion is that typical gas-particle two-phase flow moves and combustion process, the numerical simulation of Gas-solid Two-phase Flow The main simulation including gas phase turbulance of research, simulation of the gentle solid phase interphase interaction of simulation of granule motion etc., it is in list Grow up on the basis of the simulation of phase turbulent flow.Have at present two kinds of different viewpoints to the research of two phase flow: one be fluid or Gas, as continuous media, is been described by eulerian coordinate system, and using particle swarm as discrete system, in Laplace coordinate system It is been described by；And another be in addition to a fluid as continuous media outside, also particle swarm as quasi-continuous medium or fluid analogy, two Coexist in space and interpenetrate, biphase be all been described by eulerian coordinate system.Instantaneous equation to turbulent multiphase flow Group carries out after Reynolds decomposes and be average, obtaining at that time all equation group, equation according to similar to the method adopting in single-phase turbulent flow Unknown associations are contained it is impossible to close after homogenizing during group.For this reason, proposing following simulation and method for simplifying: (1) dynamics of single particles Model；(2) small slip model；(3) no sliding model (one fluid model)；(4) two-fluid model (multi-phase model or sliding Many continuum Model of diffusion)；(5) particle-trajectory model.

Particle-trajectory model, processes granule phase in Largrangian coordinates, and considers unrelated with particles diffusion, biphase Between big velocity-slip and temperature glide, taken into full account the alternate interaction of gas phase and granule.Stochastic Separated Flow Model adopts monte-carlo.Method solves the random orbit of granule motion in this random instantaneous flow field, and the granule to count fluid turbulent is made With.Particle-trajectory model is easy to simulate the experience of the granule of evaporation, volatilization and out-phase reaction, no numerical value in the forecast of granule phase Diffusion.Therefore, this model is current most widely used model in turbulent flow with combustion simulation.But it cannot meet pulsation The continuity equation of amount is it is impossible to simulate particle fluctuation completely.If obtaining the granule details compared with experimental result, need Amount of calculation that will be very big.

What this report selected is Lagrangian stochastic particle model trajectory, and a difficult problem for this model solution is concentrated mainly on three Aspect: the coupling of gas-particle two-phase, fluid velocity simulation, boundary condition.Its basic thought is: when calculating the random orbit of granule Consider the impact that gas-phase fluctuation random velocity is moved to granule, that is, by the instantaneous equation of momentum of granule, randomly give gas Instantaneous velocity, with monte-carlo method calculate random instantaneous flow field in granule random orbit with count fluid turbulent to The effect of grain.Its major advantage is to calculate simply, when granule has more complicated change through lasting, can preferably follow the trail of the fortune of granule Dynamic, also will not produce pseudo- diffusion during numerical computations；It is exactly also the impact considering fluid turbulent pulsation to granule.Its shortcoming is It is difficult to simulate the transport process of KεT model completely it is also difficult to provide granule that can be corresponding with the granule Euler field feature of actual measurement Speed and the close data of concentration space distribution.

Form (x direction) under cartesian coordinate system for the force balance equation of granule is:

$\frac{{\mathrm{du}}_p}{dt}=f_d(u-u_p)+\frac{g_x({\mathrm{\ρ}}_p-\mathrm{\ρ})}{{\mathrm{\ρ}}_p}+f_x---(2-12)$

Right formula Section 1 is stress suffered by granule, and Section 2 is granule gravity itself, and Section 3 is additional tension stress.

$f_d=\frac{18\mathrm{\μ}}{{\mathrm{\ρ}}_p{d}_p^2}\frac{c_d\mathrm{re}}{24}---(2-13)$

$f_x=\frac{1}{2}\frac{\mathrm{\ρ}}{{\mathrm{\ρ}}_p}\frac{d}{dt}(u-u_p)---(2-14)$

Wherein, u is gas phase velocity, u_pFor particle speed, μ is fluid kinematic viscosity, and ρ is gas density, ρ_pClose for granule Degree, d_pFor particle diameter.

In (2-13), re is relative Reynolds number (particle Reynolds number), and expression formula is as follows:

$\mathrm{re}=\frac{{\mathrm{\ρd}}_p|u_p-u|}{2}$

c_dFor stress coefficient, expression formula is as follows:

$c_d=a_1+\frac{a_2}{\mathrm{re}}+\frac{a_3}{{\mathrm{re}}^2}$

a₁、a₂、a₃It is ball granule constant, according to morsi and alexander et al. its value of view with re value range And change.

Or, according to the saying of haider and levenspiel, can also be expressed as

$c_d=\frac{24}{\mathrm{re}}(1+b_1{\mathrm{re}}^{b_2})+\frac{b_3\mathrm{re}}{b_4+\mathrm{re}}$

Wherein, b₁=exp (2.3288-6.4581+2.448 φ²)

b₂=0.0964+0.5565 φ

b₃=exp (4.905-13.8944+18.4222 φ²-10.2599φ³)

b₄=exp (1.4681+12.2584 φ -20.7322 φ²+15.8855φ³)

The definition of form factor φ is: φ=s/s

S is that the ball spherome surface having same volume with granule amasss, and s is the surface area of granule.In simulation, φ is set to 1, that is, It is assumed that pulverized coal particle is spheroidal particle.

Under Lagrange remainder, the instantaneous equation of momentum of granule is:

$m_p\frac{{\mathrm{dv}}_p}{dt}=\mathrm{\σ}f$

M in formula_p、v_p, t, f be respectively the quality of granule, speed, movement time and suffered power.Ignore the floating of granule Power, magnus power, effect of barometric gradient power, saffman power and false mass force etc..The equation of momentum of therefore granular mass For:

$\begin{array}{ccc}\frac{{\mathrm{du}}_p}{dt}=\frac{1}{{\mathrm{\τ}}_p}(\stackrel{\&overbar;}{u}+u^{\′}-u_p)& \frac{{\mathrm{dv}}_p}{dt}=\frac{1}{{\mathrm{\τ}}_p}(\stackrel{\&overbar;}{v}+v^{\′}-v_p)-g& \frac{{\mathrm{dw}}_p}{dt}=\frac{1}{{\mathrm{\τ}}_p}(\stackrel{\&overbar;}{w}+w^{\′}-w_p)\end{array}$

Wherein, τ_pFor granule slack time, u ', v ', w ' are the fluctuation velocities of gas phase it is assumed here that gas phase turbulance flow field is Local uniform and isotropic, when granule is located at certain turbulent vortices, u ', v ', w ' are taken as:

$\begin{array}{ccc}{u}^{\′}=\mathrm{\ζ}\sqrt{u^{\′2}}=\mathrm{\ζ}\sqrt{\frac{2}{3}k}& v^{\′}=\mathrm{\ζ}\sqrt{v^{\′2}}=\mathrm{\ζ}\sqrt{\frac{2}{3}k}& w^{\′}=\mathrm{\ζ}\sqrt{w^{\′2}}=\mathrm{\ζ}\sqrt{\frac{2}{3}k}\end{array}$

In formula,For the mean square root of gas phase turbulance fluctuation velocity, ζ is to meet Gauss to divide The random number of cloth, k is gas phase turbulance kinetic energy.

The equation of locus of granule is:

x_p=∫ u_pdt y_p=∫ v_pdt z_p=∫ w_pdt

Gas phase turbulance combustion model

The appearance of turbulent flow, not only can affect the characteristic in flow field, and influences whether all of transport equation.Turbulent combustion speed Rate is affected by turbulent flow, molecular transport and chemical kineticses three aspect simultaneously, has not yet to see the rapids commonly using Stream burn rate formula.The simulation of presently described turbulent flow gas phase combustion process, mainly have k- ε-g model for diffusion flame, For the vortex breakdown model (ebu) of premixed flame, draw cut sliding formwork type, escimo turbulent combustion theory etc..

In gas phase diffusion flame, fuel and oxidant are in various flows.Before it reacts, both connect Touch and must reach molecular level.When the incorporation time scale ratio response time, yardstick was much bigger it is necessary to consider turbulent closure scheme in detail Process, but assume that is transient chemical reaction (fast chemical reaction), can calculate course of reaction using balanced algorithm.? There are kind of the easily widely used method assumed based on this, i.e. Hybrid analysis method in comprehensive combustion model.

Hybrid analysis probability density function (pdf) model is not required to solve the transport equation of each component, only solves one The transport equation of individual or two conserved quantities (Hybrid analysis), the distribution of the concentration of the single component Hybrid analysis according to prediction is asked Solution.The impact of turbulent flow in the solution procedure of conserved quantity, is considered using probability density function.It is applied to incompressible turbulent flow Flow field and diffusion combustion response system are it is impossible to be used for premix or partly-premixed combustion system.

A kind of binary system to fuel and a kind of oxidant composition, the definition of Hybrid analysis f is represented by:

$f=\frac{z_i-z_{i,ox}}{z_{i,fuel}-z_{i,ox}}---(2-16)$

In formula: z_iElement mass fraction for element i.Subscript ox represents the value of oxidant stream porch, and fuel represents combustion The value of stream porch.

Under the hypothesis of identical diffusibility, composition equation can be reduced to a single equation with regard to blending ingredients f. F is a conserved quantity, and time average Hybrid analysis equation is:

$\frac{\∂}{\∂t}\left(\mathrm{\ρ}\stackrel{\&overbar;}{f}\right)+\▿\·\left(\mathrm{\ρ}\stackrel{\&overbar;}{u}\stackrel{\&overbar;}{f}\right)=\▿\·(\frac{{\mathrm{\μ}}_t}{{\mathrm{\σ}}_t}\▿\stackrel{\&overbar;}{f})+s_m+s_{user}---(2-17)$

Source item s_mOnly refer to quality by the incoming gas phase of reaction particles (as coal).s_userDefine source item for any user.

Except solve Hybrid analysis when equal equation in addition to, there is also a need for solving average Hybrid analysis mean-square valueConservation equation:

$\frac{\∂}{\∂t}\left(\mathrm{\ρ}\stackrel{\&overbar;}{f^{\′2}}\right)+\▿\·\left(\mathrm{\ρ}\stackrel{\&overbar;}{u}\stackrel{\&overbar;}{f^{\′2}}\right)=\▿\·(\frac{{\mathrm{\μ}}_t}{{\mathrm{\σ}}_t}\▿\stackrel{\&overbar;}{f^{\′2}})+c_g{\mathrm{\μ}}_t\left({\▿}^2\stackrel{\&overbar;}{f}\right)-c_d\mathrm{\ρ}\frac{\mathrm{\ϵ}}{k}\stackrel{\&overbar;}{f^{\′2}}+s_{user}---(2-18)$

In formula:Constant σ_t、c_g、c_dTake 0.85,2.86,2.0 respectively.

Hybrid analysis analogy method benefit is the Hybrid analysis that chemical reaction is reduced to one or two conservation.Institute There is heat chemistry scalar (constituent mass fraction, density and temperature) all uniquely relevant with Hybrid analysis.Given reaction coefficient chemical Matter and chemical reaction, in flow field, the instantaneous conservation fractional value of any point can be used for calculating each component molar fraction, density And temperature value.Just can simulate in the middle of in turbulent reacting flows because Hybrid analysis method need not solve substantial amounts of mass transfer equation The interaction of the formation of product, turbulent flow and chemical reaction, and its computational efficiency is high, more accurate than finite rate method can draw Average fluid density, so be widely used at present.

Coal dust firing model

Process of coal combustion is divided into several stages: pulverized coal preheating, fugitive constituent separate out and combustion process, coke burning waited Journey.The fugitive constituent producing during pyrolysis of coal and the burning of fugitive constituent have important impact for whole burning of coal process, sometimes Even conclusive impact.In order to obtain the differential equation group closed, the source item master in energy equation and component balanced equation Jointly to be given by volatile matter devolatilization model and coke granule combustion model.

(1) fugitive constituent precipitation model

There is pyrolytic reaction in coal, separate out fugitive constituent, the then volatilization of remaining coke and precipitation at high temperature first Part is reacted with the oxygen of in the air respectively:

Raw coal: y fugitive constituent+(1-y) coke

On the basis of numerical computations test, researcher establishes different devolatilization model.Mainly there is fixed volatilization speed Rate model, single reaction model, two-equation model, multistep parallel reaction model, functional group's devolatilization model, consideration non-dynamics control Devolatilization model of factor etc..

By stickler et al. in 1975 propose double parallel reaction model be currently apply more extensive model, should Model is thought rate constant

With

In the model, e₂＞ e₁, a₂＞ a₁.In lower temperature, first reaction plays a major role, during high temperature, second Reaction plays a major role.

Discharge fugitive constituent ratio and be respectively α₁And α₂, control activation energy numerical value so that first reaction is carried out at low temperature, the Two reactions are carried out at high temperature, and the Mass lost amount of the coal caused by pyrolysis is:

$v=\underset{0}{\&integral;}\{({\mathrm{\α}}_1\·k_1+{\mathrm{\α}}_2\·k_2)\·\exp \mathrm{\[}{\&integral;}_0^t(k_1+k_2)dt\]\}dt---(2-21)$

Fugitive constituent mass change:

$\frac{dv}{dt}=-(y_1{k}_1+y_2{k}_2)\·m_c---(2-22)$

Wherein: m_cFor unreacted raw coal quality；a₁,a₂,e₁,e₂For fugitive constituent pyrolysis Kinetics Parameter, by measuring.By Simple in calculating, result of calculation has certain accuracy, therefore two-equation model apply in realistic simulation very wide.

(2) combustion simulation of coke

In the combustion process of pulverized coal particle, the burning of coke is a complicated process (belonging to gas-solid heterogeneous combustion), This process include oxidation material to the diffusion process of particle surface and these materials particle surface and coke course of reaction, one As think, this two processes are carried out under a quasi-equilibrium reactive state simultaneously.Coke granule inhomogeneous reaction process Simulation is complicated, it be subject to the fragmentation of coke, endoporus diffusion, change of the change of surface area and temperature and pressure etc. these The impact of uncertain factor.

Because the burning of coke is not only relevant with diffusion, also relevant with kinetics factor.Thus coke burning at present Modal in the model using is exactly eddy diffusion combustion model.Burn rate r of coke_cIt is diffused into coke table by oxygen Speed k in face_dChemical reaction rate k with coke surface_cThe co- controlling of these two aspects, is represented by:

$r_c=\frac{6r_2{p}_{o_2}}{d}\frac{1}{(1/k_c+1/k_d)}---(2-25)$

$\begin{array}{cc}{k}_c=a_c{e}^{-e_c/{\mathrm{rt}}_2}& k_d=s_h{d}_{o_2}{m}_c/{\mathrm{rt}}_{1}d\end{array}$

In formula, p_o2Partial pressure for oxygen；d_o2Diffusion coefficient for oxygen；s_kFor sherwood number, it is taken as 2.0；Mc is carbon Molal weight.

In the present embodiment, fugitive constituent separates out and uses two-equation model, and coke burning uses diffusion kinetics Control combustion model.In a model it is assumed that coal grain is the spheroidal particle of single particle size, in any moment of reaction, coal grain is by water Divide, raw coal, coke and this four part of ash form.The rising of coal grain temperature leads to the moisture evaporation in coal grain, enters gas phase and becomes Become water vapour.Raw coal separates out with fugitive constituent and consumes, and remaining solid combustible is coke, and coke and oxygen occur out-phase anti- Should and gradually after-flame.Ash gradually trends towards 1 with the after-flame of coke.The composition of the fugitive constituent that hypothesis separates out is nytron Thing (ch_x), it exhausts rapidly in gas-phase combustion reaction.

Radiative heat transfer model

Radiant heat transfer process is the main heat transfer type in high-temperature burner hearth, therefore in Combustion simulation system, to radiation The simulation of the transmission of energy is extremely important, is also an extremely complex task simultaneously.Such as in a typical coal dust firing In stove, radiation includes granule, coal dust, coke, soot particle, bituminous coal and gas phase (mainly co simultaneously₂And h₂O) effect.Radiation calculates Degree of accuracy depend on the levels of precision of adopted computational methods and the degree of understanding for radiation medium and surrounding wall property. Currently for different applicable elements, have been developed in the computation model of a lot of radiation heat transfers, the method master that these models adopt Heat flow method (heat flux) to be had, Monte Carlo method (monte-carlo) and region analysis method (zone analysis).

The feature of heat flow method is by complicated uneven multinomial interface radiant heat flux uniform interface radiant heat flux To replace, and to average.The advantage of heat flow method is to calculate simple and amount of calculation is little.But for having the area of intense radiation Domain, the hypothesis of heat flow method is clearly present in true discrepancy.But because the method calculates easy, the model that error allows in engineering In enclosing, therefore apply more at present in boiler flow field numerical simulation, development is more ripe perfect.

The model that this report selects is the p-1 radiation patterns based on heat flow method.P-1 method is that a simplest bulb is humorous Function method, it assumes that the radiant intensity in medium along the orthogonal thereto spheric harmonic function of space angle be distributed, and by containing differential, integration Emittance equation of transfer is converted into one group of partial differential equation, and simultaneous energy equation and corresponding boundary condition just can obtain spoke Penetrate the spatial distribution of intensity and temperature.Compared with do method, p-1 method considers the effect of radiating scattering, is more suitable for optical thickness Combustion apparatus big and that geometry is complicated, and the time solving required for emittance equation is short, is relatively suitable for solving coal Burn in powder stove.Researcher both domestic and external this model multiplex when simulating coal-powder boiler burning.

For radiant heat flux q_r,

$q_r=-\frac{1}{3(\mathrm{\α}+{\mathrm{\σ}}_s)-{\mathrm{c\σ}}_s}\▿g---(2-26)$

Wherein: α is absorptance, σ_sFor scattering coefficient, g is incident radiation, and c is linear anisotropic phase function system Number.Introducing parameter:

$\mathrm{\γ}=\frac{1}{3(\mathrm{\α}+{\mathrm{\σ}}_s)-{\mathrm{c\σ}}_s}$

Equation can turn to:

$q_r=-\mathrm{\γ}\▿g$

The transport equation of g is:

$\▿(\mathrm{\γ}\▿g)-\mathrm{\α}g+4{\mathrm{\α\σt}}^4=s_g---(2-27)$

Wherein: σ is Stefan Boltzmann constant, s_gFor user-defined radiation source phase.During using p-1 model, ask Solve this equation to obtain local radiant intensity.

Merge two formula above, can get equation below:

$-\▿q_r=\mathrm{\α}g-4{\mathrm{\α\σt}}^4---(2-28)$

Expression formula can be brought directly to energy equation, thus obtaining due to the caused calorie source (remittance) of radiation.

The impact of granule when including granule dispersion phase in model, can be considered in p-1 radiation patterns.For comprising There is the grey body medium of absorption, transmitting, scattering nature granule, the transport equation of incident radiation is:

$\▿(\mathrm{\γ}\▿g)+4\mathrm{\π}(\frac{{\mathrm{\α\σt}}^4}{\mathrm{\π}}+e_p)-(\mathrm{\α}+{\mathrm{\α}}_p)g=0---(2-29)$

Wherein, e_pFor the radiation-emitting amount of equal value of granule, α_pEquivalent absorption coefficient for granule.

$\begin{array}{cc}{e}_p=\underset{v\→0}{\lim }\underset{n=1}{\overset{n}{\σ}}{\mathrm{\ϵ}}_{pn}{a}_{pn}\frac{{\mathrm{\σt}}_{pn}^4}{\mathrm{\π}v}& a_p=\underset{v\→0}{\lim }\underset{n=1}{\overset{n}{\σ}}{\mathrm{\ϵ}}_{pn}\frac{a_{pn}}{v}\end{array}$

Wherein, ε_pn, a_pn, t_pnIt is respectively blackness, projected area (vertical radiation direction) and the temperature of n-th granule.

The definition of γ is:

$\mathrm{\γ}=\frac{1}{3(\mathrm{\α}+{\mathrm{\α}}_p+{\mathrm{\σ}}_p)}$

Wherein, equivalent particle dispersion factor is defined as:

${\mathrm{\σ}}_p=\underset{v\→0}{\lim }\underset{n=1}{\overset{n}{\σ}}(1-f_{pn})(1-{\mathrm{\ϵ}}_{pn})\frac{a_{pn}}{v}$

It obtains during granule following calculation, f_pnScattering coefficient for n-th granule.

Radiant heat strength q in p-1 method_rWall boundary condition be:

$q_{r,w}=-\frac{4{\mathrm{\π\ϵ}}_w\frac{{\mathrm{\σt}}_w^4}{\mathrm{\π}}-(1-{\mathrm{\ρ}}_w)g_w}{2(1+{\mathrm{\ρ}}_w)}---(2-30)$

If it is assumed that wall is diffusion grey body surface, then ρ_w=1- ε_w, above formula can be expressed as:

$q_{r,w}=\frac{4{\mathrm{\ϵ}}_w}{2(2-{\mathrm{\ϵ}}_w)}(4{\mathrm{\σt}}_w^4-g_w)---(2-31)$

This equation is used for calculating the q in energy equation_r,wAnd the boundary condition of radiation equation.

no_xFormation mechanism and model

(1)no_xFormation mechanism

The no generating in coal combustion process is thought in research at present_xIt is divided into three types, respectively heating power type, Quick-type and combustion Material type.No in impact burning_xThe factor generating has fuel characteristic such as coal, nitrogen content, nitrogen substance structure, grain diameter etc.； Service condition such as boiler form, load, temperature, oxygen amount, response time (time of staying) etc..

(2) heating power type no_x(thermal no_x) formation mechanism

Heating power type no_xRefer to the n in combustion air₂The nitrogen oxides aoxidizing at high temperature and generating.Its formation mechanism It is that zeldovich proposed in nineteen forty-six, its generating process can be described by following reaction:

When fuel concentration is excessive in addition it is also necessary to consider following formula reaction:

Heating power type no_xHigh-temperature region mainly more than 1500 DEG C produces, the relation of its formation speed and temperature be by According to Ah 's Li Wusi law: with the rising of temperature, no_xFormation speed exponentially increase sharply.Its growing amount is permissible Estimated by zeldovich kinetic model:

$\frac{d\[no\]}{dt}=3\×{10}^{14}\[n_2\]{\[o_2\]}^{1/2}\exp (-542000/rt)---(2-35)$

In formula: [o₂], [n₂], [no] is respectively o₂, n₂, the concentration (mol/cm of no³)；

T is absolute temperature (k)；

T is the time (s)；

R is universal gas constant (j/ (mol k)).

Heating power type no_xFormation speed have exponent relation with temperature, when ignition temperature be less than 1800k when, heating power type no_xRaw Become few, when temperature is higher than 1800k, reaction is gradually obvious, and the rising with temperature, no_xGrowing amount drastically raise, Temperature in 1800k about when, temperature often raises 100k, and reflection speed will increase 6-7 times.In combustion, if there is office Portion high-temperature region, then can generate more no in these regions_x, it may be to the whole indoor no that burns_xGenerate key work With should avoid the effect in area of localized hyperthermia therefore in real process as far as possible.

Excess air coefficient is to heating power no_xImpact be also obviously, in theory for heating power no_xGrowing amount and oxygen The square root of concentration is directly proportional, and that is, oxygen concentration increases the concentration of oxygen atoms that oxygen molecule can be made at relatively high temperatures to decompose gained and increases Plus, make heating power no_xGrowing amount increase.And situation can be more more complicated in real process, because excess air coefficient increases On the one hand oxygen concentration, another aspect o are increased₂The increase of concentration can make ignition temperature decline due to the diluting effect of oxygen, leads to no_xGenerating rate reduce.In a word, no_xGenerating rate and o₂Concentration has the relation of an extreme value, and this extreme value is α in theory =1, deviateing α=1 all can make no_xGrowing amount reduce.

(3) Quick-type no_xFormation mechanism

Quick-type no_xIt is found through experiments in 1971 by Fenimore (fenimore) earliest.In rich hydrocarbonization On the flame front face of compound fuel combustion, the substantial amounts of quick no of generation can be reacted_x, the ch that gone out due to hydrocarbon fuel pyrolytic from N by base and in the air₂Reaction generates hcn and n, and then in o₂In the presence of form no at a terrific speed_x, required for reaction Time be about 60ms, 0.5 power of growing amount and furnace pressure is directly proportional, little with the relation of temperature.Due to quick no_x Hydrocarbon is needed to start and n₂Reaction, so growing amount is more in fuel-rich flame, the multiple combustion being born in internal combustion engine Burning process, for coal combustion equipment, Quick-type no_xOnly account for 5% about.

(4) fuel type no_xFormation mechanism

Fuel type no_xThe no generating in coal dust firing_xIn account for very big ratio.Either fugitive constituent burning or coke combustion The burning stage all defines substantial amounts of no_x, coal fuel type no during pyrolysis and combustion_xGeneration and reduction process extremely complex, it With the ratio in fugitive constituent and coke, nitrogen-containing products composition and combustion after coal characteristic, the structure of nitrogen content of coal, nitrogen decomposes Burning condition is closely related.Burning type no_xGeneration reduction process substantially can divide following three steps: (a) pyrolysis devolatilization process: wave Send out the precipitation of part n；(b) oxidizing process: the course of reaction of fugitive constituent n and coke n and in the air oxygen；(c) double competitive reaction process: The nitrogenous intermediate product of fuel n conversion generates no_xOxidation reaction and generate no_xIt is reduced into n containing oxygen intermediate₂Also The process that former reaction occurs simultaneously and vies each other.

(5)no_xGeneration model

No in combustion process_xGenerating process and its complexity, mainly have following no at present_xGeneration model: elementary reaction De ' soete model of model, pdf transport equation model, de ' soete model and extension etc..

(6) elementary chemical reaction model

Elementary chemical reaction model have ignored the feature during Actual combustion, emphasizes no_xThe balancing response generating is it is considered to system Interior all possible elementary reaction and reacting substance, are solved by equilibrium equation.Elementary chemical reaction model precision is higher, but due to being related to Reaction a lot, calculate more complicated.

(7) de ' soete model

De ' soete model is set to hcn all intermediate products, in coal the rate of release of n be proportional to pulverized coal pyrolysis and Mass attenuation rate during coal char combustion.

$r_{coal\→hcn}=2s_p{f}_n\frac{m_{hcn}}{m_{n_2}}$

Wherein, f_nMass fraction for nitrogen content of coal；m_hcn,It is respectively hcn, n₂Molecular weight.

(8) de ' the soete model extending

De ' soete model only accounts for fuel no_xThe no that the fugitive constituent hcn of the inside produces_xWith heating power type nox, do not consider Nh in fugitive constituent₃The no generating etc. component and coke.

(9) pdf transport equation model

no_xIn generating process, not only relevant with the chemical reaction mechanism of itself, and and Turbulent Flow Field and its between Dependent interaction is relevant.Xiang Douyong closing form related with chemical reaction rate for turbulent transport expressed by pdf transport equation model, Need not simulate, any complicated chemical reaction mechanism can accurately calculate, and therefore it is in the simulation of pollutant generating process It is used widely in journey.

No in process of coal combustion_xThe growing amount of pollutant is few, mainly no, and it will not be to local other mean fields Variable, such as temperature, speed, key component concentration produce large effect.Therefore, the numerical simulation calculation that no generates can be from meter Calculate in stove and decouple out in the program of Gas-solid Two-phase Flow, heat transfer and coal dust firing, that is, adopt postpositive disposal (post- Processor method) is simulated very little, typically can ignore.Therefore, consideration is not given in the reaction to Quick-type no herein.

No in coal-powder boiler_xGenerating process betides turbulent gas particle flows disorder of internal organs, and during its turbulent flow, equal reaction rate is different from and changes The instantaneous reaction rate that mechanism model is given is it is therefore necessary to consider the impact to its chemical reaction rate for the turbulence pulsation.Coke The reduction reaction of no is reacted for out-phase, speed is slower, can ignore the impact to its chemical reaction rate for the turbulence pulsation, at that time All reaction rates directly can be calculated by average variable.Rate of release for hcn in coal can also ignore turbulence pulsation Impact.However, generating and each homogeneous reaction in reduction process for no, during due to its response time yardstick and turbulent closure scheme Between yardstick be the same order of magnitude it is therefore necessary to consider the impact to its chemical reaction rate for the turbulence pulsation.At present, apply and must compare Universal method is using the impact setting pdf (probability density function) model method simulation turbulent flow, Compared with pdf transport equation model method, this model is difficult to simulate the phase between detailed limited chemical kineticses and turbulent flow Interaction, but for no_xThe general simulation generating is suitable, and amount of calculation is less, and has higher computational accuracy.This Research sets the function as two variable-temperature, oxygen concentration for the instantaneous reaction rate, using the setting beta function shape of limited response speed The pdf model of formula, i.e. p (t), p (y_o2), instantaneous reaction rate is multiplied by p (t) p (y_o2) be integrated it is possible to draw average Reaction rate.It is θ that temperature t is carried out nondimensionalization, supposes θ and y simultaneously_o2It is the variable of two statistical iteration, concrete model is:

${\stackrel{\&overbar;}{s}}_{no}=\&integral;\&integral;s_{no}p(\mathrm{\θ},y_{o_2}){\mathrm{d\θdy}}_{o_2}=\&integral;\&integral;s_{no}p\left(\mathrm{\θ}\right)p\left(y_{o_2}\right){\mathrm{d\θdy}}_{o_2}---(2-36)$

Method of value solving

By to the conservation of mass in combustion process, preservation of energy, the conservation of momentum, component balanced and kinetics base The research of this rule, establishes flowing in combustion process, the mathematical model of heat transfer, mass transfer and combustion phenomena and each subprocess. The governing equation of combustion process is complicated non-linear partial differential Fang Yicheng, in addition to indivedual simple case, is difficult to directly solving Method obtains the accurate solution of these partial differential equation, can only seek its approximate solution with iterative method, therefore, sets up correct physical model Afterwards, it is critical only that whether can set up suitable method for solving.

(1) discretization in region

Before carrying out numerical simulation to boiler combustion, first have to carry out the discretization of zoning, its essence is exactly Replace original continuous space with limited discrete point, i.e. gridding.Before the discretization in region is by numerical computations Carry.The method of gridding is a lot, has fixed mesh and floating grid, orthogonal grid and non-orthogonal mesh, be uniformly distributed grid and Non-uniform Distribution grid, staggered-mesh etc..

According to dependent variable in the distribution supposition of node and the method for derivation discrete equation, mainly there is following several discrete side Method: finite difference calculus, FInite Element and finite volume method.Use finite volume method (finite volume herein Method abbreviation fvm) also known as control volume mothod.The basic ideas of finite volume method: zoning is divided into grid, and makes every There is a mutually unduplicated control volume around individual mesh point, the differential equation to be solved is controlled volume integral to each, thus Draw one group of discrete equation.From the choosing method of integral domain, finite volume method belongs to the SUBDOMAIN METHOD in weighted residual method, From the approximation method of Unknown Bound, finite volume method belongs to the discrete method using Local approximation.

Additionally, regarding position in controlling volume for the node, interior nodes method and exterior node method can be divided into.The mode shadow of gridding Ring the discrete difficulty or ease of the differential equation.Also it is related to the precision of solution, convergence and economy

(2) several discrete scheme

When setting up discrete equation using finite volume method it is important that a step be exactly will control volume interface on physics Amount and its derivative are obtained by node physical quantity interpolation.Mainly there is following several discrete scheme: central difference schemes, single order at present Upstreame scheme, mixed format, exponential scheme, power law scheme, Second-order Up-wind form and quick form.

Central difference schemes (central differencing scheme) are it is simply that the physical quantity on interface is using linear Formula for interpolation is calculating.Central difference schemes can be only applied to speed very little or mesh spacing very little.

Monovalence upstreame scheme (first order upwind scheme) considers the impact of flow direction, at any All without the concussion causing solution under part, forever can obtain physically it appear that suitable solve.Therefore, single order upstreame scheme obtains To being widely applied.

The factor that mixed format (hybrid scheme) combines centered difference and acts on two side's one sides windward.Therefore, mix Qualified formula is widely adopted in cfd software.Shortcoming is only single order precision.

Exponential scheme (exponential scheme) by spread with the effect of convection current and together with considering.Exponent arithmetic Can accurately be solved, but computing expense is time-consuming, in the problem of two and three dimensions, and the situation that source item is not zero, calculate not Accurately.

Power law scheme (power-law scheme) and a kind of exponential scheme discrete scheme closely.

Second-order Up-wind form and single order upstreame scheme are it is identical in that all determine control by the physical quantity of upstream units point The physical quantity at volume interface processed.But Second-order Up-wind form not only uses the value of a nearest point of upstream, other one also to be used The value of individual point upstream.

Quick (quadratic upwind interpolation of connective kinematics) form is A kind of method improving discrete equation intercept.

This report adopts monovalence upstreame scheme.

(3) flow field numerical algorithm

The basic process of Flow Field Calculation is spatially with finite volume method, computational fields to be separated into many little volume lists Unit, solves to the governing equation group after discrete in each elementary volume, volume element.Coupling type solution and separate type solution can be divided into Method, wooden literary grace separate type solving method.Separate type solving method is sequentially, seriatim to solve each equation (with regard to u, v, w, p and t Equation).After namely first an equation (as the u equation of momentum) being solved on total-grid, then solve another one equation (as v The equation of momentum).Because governing equation is nonlinear, and it is coupling each other, therefore, before obtaining convergence solution, Take turns iteration through excessive.Each wheel iteration is made up of following steps: the 1, result according to current solution, updates all flowing momentum, such as Fruit calculates and just starts, then updated with initial value；2nd, solve u, v and w equation of momentum respectively in order, obtain velocity field, note When being intended to calculate, the mass flow of pressure and unit interface uses current given value；3rd, because the speed that the 2nd step obtains is likely to It is unsatisfactory for continuity equation, therefore, construct the pressure correction side of a poisson type with continuity equation and the linearizing equation of momentum Journey, then solves this pressure correction equation, obtains the correction value of pressure field and velocity field；4th, using the velocity field newly obtaining and pressure The field of force, solves the governing equation of other scalars (such as temperature, tubulence energy and group are graded)；5th, for the simulation comprising discrete phase, when When inside has Coupling Between Phases, the trajectory calculation result according to discrete phase updates the source item of continuous phase；6th, whether check equation group Convergence.If not restraining, returning to the 1st step, repeating.

Simple (semi-implicit method for pressure-linked equations) is current engineering In a kind of wide variety of Flow Field Calculation method, it belongs to one kind of pressure-type cable.Mainly include simple and derived from by it Simpler, simplest and piso out.

The present embodiment adopts simple algorithm.

Stove combustion process numerical simulation

(1) calculate grid

Turbulent burner is single only tissue burning, and therefore the burner of swirl-opposed boiler has very big shadow to boiler combustion Ring.Because burner is for boiler, size is comparatively small, and the grid of burner region can be very big, and this is to computer Performance requirement is very high.Have no effect on coal dust firing in burner hearth in view of the process in horizontal flue and back-end ductwork, and coal dust Grain moves to burning during flue to be terminated substantially, and flue gas does not substantially occur chemical reaction in flue, does not therefore have herein To the process simulation in back-end ductwork, and have ignored the impact of superheater stream field in horizontal flue.According to upper one section The flow behavior of burner outlet, as boiler-burner entrance boundary condition, builds domain and carries out stress and strain model, Furnace outlet increases the horizontal segment of 10m, to obtain fully developed flow in exit.In order to obtain high-quality calculating grid, adopt Use multiblock technique division methods, using structuring hexahedral mesh, for accurate simulation burner outlet aerodynamic field, burn Device entrance area needs to carry out grid local cypher, and total grid number is 1,500,000.Chamber structure and grid such as Fig. 3 (a) to (c) Shown.

Simulated condition and computational methods

Burner body is simplified, builds burner inlet model according to burner dimensions；Centre wind and First air Using speed entrance boundary condition, inner second air and outer second air adopt quality entrance boundary condition；Porch wind speed, pathogenic wind-warm, Mass flow is configured according to the burner outlet flow behavior of design parameter, variable working condition condition and a upper section.To after-flame Wind and side burnout degree body are simplified, and build its entrance model according to burner dimensions；Burnout degree and side burnout degree adopt matter Amount entrance boundary condition, mass flow values are calculated according to design parameter and variable working condition condition.In the condition of entrance boundary, The definition of velocity component adopts partial cylindrical coordinate system, sets the velocity component of each spout respectively.Export boundary condition adopts Pressure export, pressure is set to -80pa；Burner hearth wall adopts standard law of wall equation, no slip boundary condition, and heat exchange adopts the Two class boundary conditions, i.e. temperature boundary condition, given wall surface temperature and radiance, wall surface temperature is 700k, and Wall Radiation rate is 0.6.It is contemplated that upper furnace screen superheater and high temperature superheater must affect on afterbody flow field and heat exchange in simulation.

Pulverized coal particle diameter is distributed according to rosin-rammler method.Rosin-rammler distribution supposes in particle diameter The d and mass fraction y of the granule more than this diameter_dBetween there is exponential relationship:

$y_d=e^{-{(d/\stackrel{\&overbar;}{d})}^n}$

For average diameter, n is profile exponent

The coal dust sample that phenomenon is obtained carries out laser particle size analysis, obtains coal particle size distributed data: minimum grain size 5 μ M, 250 μm of maximum particle diameter, 63.8 μm of mean diameter, profile exponent 1.008.

Carry out cold conditions calculating first, then carry out hot calculating again, until convergence.Pressure for discrete equation group and speed Degree coupling adopts simple Algorithm for Solving, and solving equation adopts by-line iterative method and underrelaxation factor, and convergence is energy, spoke Penetrate heat transfer, no and hcn calculates residual error and be less than 10^-6, remaining items calculating residual error is less than 10^-3.

Rated load design conditions sunykatuib analyses

First boiler rated load design conditions are simulated, load 600mw, 6 coal pulverizers run, and run oxygen amount 2.6%, each burner output and air distribution uniform.

Velocity field

Fig. 4 is burner y=9.56m section speed vector figure, fully reflects burner and liquidates the feature of arrangement, air-flow After colliding in burner hearth center, the overwhelming majority flows up, and an orlop burner part flows down to furnace hopper, forms whirlpool Area；The speed of ascending air is 8-15m/s, air-flow in burner hearth degree of filling very well, be reasonably distributed, with the increasing of furnace height Plus, speed incrementally increases.

Fig. 5 is the speed field pattern in middle level burner and ofa region.It can be seen that First air is in burner hearth There is stronger rigidity, but be not in the situation that flame directly impacts opposite water-cooling wall, combustion zone has larger backflow Area.Middle part air-flow has the situation pressing to side wall, and whirlpool in burner hearth corner.In ofa section, some are inclined for burnout degree rigidity Weak it is difficult to blow in the middle part of burner hearth.

Fig. 6 is away from front-back wall burner outlet 0.1m speed vector figure.In figure air-flow has one significantly to rotate, each burning Having asked of device significantly interacts, and because adjacent burners reversely rotate, the composite result of swirling eddy gas phase velocity makes gas The tangential velocity of stream increases, and has a common exteenal reflux area between turbulent burner.

Fig. 7 be middle level burner and ofa region temperature cloud picture.It can be seen that because turbulent burner is Single only tissue burning, therefore in burner hearth, flue-gas temperature is generally symmetric, and levels Temperature Distribution feature is identical, in combustion The temperature rate of change in burner exit is larger, temperature rise fast it is ensured that the catching fire in time of breeze airflow, burn in burner hearth central area The most violent, peak temperature reaches 1700 DEG C about, and furnace flame degree of filling is good.

Fig. 8 is y=9.56m section temperature cloud atlas, and burner hearth high-temperature region is located at burner to burnout degree region, and burnout degree plays The effect of banking fire, also has one section of high-temperature region after burnout degree, this is to be reacted due to the oxygen that unburnt co and c is supplemented with burnout degree. Along furnace chamber width, burner hearth central temperature is evenly distributed, and this is beneficial to reduce wall thermic load deviation.

Away from burner outlet 0.3m temperature cloud picture, there is a recirculating zone, backflow temperature between First air and secondary wind in Fig. 9 Close to 1000 DEG C, the temperature of secondary wind periphery, more than 1000 DEG C, illustrates the good flammability of burner to degree, and breeze airflow enters Can catch fire rapidly after entering burner hearth, increased pulverized coal particle in the time of staying of high-temperature region it is ensured that coal dust is lighted rapidly and filled Divide after-flame.

Figure 10 is the concentration distribution cloud atlas of smoke components on the vertical section that in stove, y=9.56m burner is located.Figure 11 A () to (c) is o₂、co、no_xThe concentration distribution cloud atlas of smoke components on the section that middle level burner and burnout degree are located.From figure In visible in burner hearth center o₂Concentration is very low, and with the increase of furnace height, oxygen concentration is integrally on a declining curve, but every Layer burner regional area, due to supplementing air, has the fluctuation of an increase, in furnace outlet region, oxygen concentration regional balance. Because burner region excess air coefficient is 0.8, therefore define a strong reducing property atmosphere area in burner region, not only Substantial amounts of co generates, also part co₂It is reduced, the oxygen that co and burnout degree supplement after burnout degree burns away generation co₂；But because the penetrance of burnout degree is weaker, in the still higher co concentration of furnace outlet, it is higher that this shows that boiler exists Loss of the inadequacy burning for chemistry mechanics reason.

Knowable to the formation mechanism of nitrogen, nox concentration is relevant with the nitrogen content of excess air coefficient, ignition temperature and coal, figure Show it can be seen that nox is consistent with temperature changing trend.Figure below is middle level burner and the vertical of y=9.56m burner place cuts Nox concentration distribution cloud atlas on face.As can be seen from the figure in burner outlet region, due to thermal-flame radiation and backflow flue gas Volume is inhaled substantial amounts of hot-air and is mixed with coal dust, and the temperature of pulverized coal particle improves rapidly, and volatile matters separate out in a large number, and nitrogen compound is rapid By a large amount of dioxygen oxidations, generate the intermediate products such as hcn, then aoxidize nox further, also can react with nox rapidly simultaneously, make Nox concentration reduce, in the middle part of burner hearth nox concentration reach minimum.Reduced zone between burner and burnout degree for the nox concentration continues Continuous reduction.Analyze its reason, consume substantial amounts of oxygen in primary zone coal dust firing, oxygen concentration is very low, form fuel rich area, A part of nox is reduced；The subsequently burning-out zone above after-flame wind combustor, the air needed for burnout degree aftercombustion, promote Enter last after-flame carbon and co burning, some increases of nox concentration.

Figure 12 is the pulverized coal particle movement locus figure of different-grain diameter.It can be seen that the coal dust of orlop burner Due to being affected by air-flow, a part of coal dust particle can enter furnace hopper it is possible to cause uncombusted loss and cold ash to particle Bucket slagging scorification；After the coal dust particle of middle level burner enters burner hearth, due to being affected by air-flow, coal dust particle particularly upper strata combustion After burner sprays into burner hearth, some coal dust particle can be affixed to water-cooling wall due to high temperature reflux flue gas, and herein coal dust oneself through combustion Burn temperature higher, and oxygen content relatively low (see Fig. 8 and Fig. 9), if the relatively low coal of burning meltbility, should be noted that upper strata is fired The problem of region slagging scorification between burner region and burnout degree.Observe the track spraying into granule near side wall two row's burner to understand, Part pulverized coal particle, in the presence of air-flow, can clash into water-cooling wall；Even so, but due to this regional temperature not high (see figure 7), therefore, side water wall will not occur serious slagging scorification.But, can be seen by the oxygen concentration cloud atlas of Figure 11 (a) to (c) Go out, in the middle part of burner region side wall, oxygen concentration is very low, assume consumingly reducing atmosphere, easy precocity high temperature corrosion.

Wall heat flux

Figure 13 is the radiant heat flux of You Qiang, Hou Qiang and left wall.As seen from the figure, in front and back wall surface thermic load with furnace height Increase be stepped up, between upper strata burner and burnout degree, region reaches maximum, then and slowly reduces.Along burner hearth width The increase that degree direction is also as width is stepped up, and reaches maximum in burner hearth central area.The thermic load of side wall is in after-flame On wind, region does not reach maximum.This thermic load characteristic is related to chamber flue gas temperature distribution and radioactive particles concentration distribution 's.In-furnace temperature is to increase with furnace height increase, and portion reaches peak on the burner, and granule density also reaches maximum Value, therefore, this area radiation hot-fluid is maximum, and wall thermic load is maximum.

Efficiency of combustion and nox emission performance

Boiler combustion efficiency and nox discharge capacity are counted, result is as follows: the total burn-off rate of coal dust is 99.66%, flying dust Combustible content is 1.97%, nox discharge capacity is 281mg/m³(6%o₂).This shows, this type boiler when using design coal, Unburned combustible in fly ash is higher, and nox discharge capacity is relatively low.

In the present embodiment, by multiblock technique division is carried out to preset boiler-burner model；By basal conservation eqution, Turbulent flow model, gas-particle two-phase flow movable model, gas phase turbulance combustion model, pulverized coal particle combustion model, radiation heat transfer Model and nox generation model carry out mould to the combustion process of the boiler coal-ash in the preset boiler-burner model after stress and strain model Intend；According to the result after simulation determine the airflow field in boiler-burner, the temperature field in boiler-burner, in boiler-burner Smoke components distribution, movement locus in boiler-burner for the pulverized coal particle, boiler-burner Wall Radiation hot-fluid, boiler combustion Burn efficiency and no_xDischarge capacity, solves at present because size of burner hearth is increasing, and the testing stand of manufacture full scale model is not cut Reality, the technical problem that burning, flowing, heat transfer global regularity characteristic in stove cannot be measured being led to.

Refer to Fig. 2, a kind of power boiler breeze combustion characteristics provided in an embodiment of the present invention generates the simulation dress of numerical value The embodiment put includes:

Division unit 201, for carrying out multiblock technique division to preset boiler-burner model；

Unit 202 reprocessed by model, for simplifying to the described preset boiler-burner model after stress and strain model, and Get the design parameter after setting and working condition.

Analogue unit 203, for by basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas Phase turbulent combustion model, pulverized coal particle combustion model, radiative heat transfer model and nox generation model are to described pre- after stress and strain model The combustion process putting the boiler coal-ash in boiler-burner model is simulated；

Determining unit 204, for determining the airflow field in described boiler-burner, described boiler according to the result after simulation Smoke components distribution in temperature field in burner, described boiler-burner, pulverized coal particle are in described boiler-burner Movement locus, described boiler-burner Wall Radiation hot-fluid, boiler combustion efficiency and no_xDischarge capacity.

In the present embodiment, solve at present because size of burner hearth is increasing, the testing stand of manufacture full scale model is not Correspond to reality, the technical problem that burning, flowing, heat transfer global regularity characteristic in stove cannot be measured being led to.

The above, above example only in order to technical scheme to be described, is not intended to limit；Although with reference to front State embodiment the present invention has been described in detail, it will be understood by those within the art that: it still can be to front State the technical scheme described in each embodiment to modify, or equivalent is carried out to wherein some technical characteristics；And these Modification or replacement, do not make the essence of appropriate technical solution depart from the spirit and scope of various embodiments of the present invention technical scheme.

Claims (10)

1. a kind of power boiler breeze combustion characteristics generates the analogy method of numerical value it is characterised in that including:

Multiblock technique division is carried out to preset boiler-burner model；

By basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas phase turbulance combustion model, coal dust Particle burning model, radiative heat transfer model and nox generation model are in the described preset boiler-burner model after stress and strain model Boiler coal-ash combustion process be simulated；

Airflow field in described boiler-burner, the temperature field in described boiler-burner, institute are determined according to the result after simulation State the smoke components distribution in boiler-burner, the pulverized coal particle movement locus in described boiler-burner, described boiler fires Burner Wall Radiation hot-fluid, boiler combustion efficiency and no_xDischarge capacity.

2. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that right Preset boiler-burner model also includes after carrying out multiblock technique division:

Described preset boiler-burner model after stress and strain model is simplified, and gets the design parameter after setting and work Condition condition.

3. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that institute Stating turbulent flow model is realizable k- ε two-equation model；

Wherein, k equation is

ε equation is

Gk represents that the Turbulent Kinetic being caused due to average velocity gradient is produced, and gb is to produce for the Turbulent Kinetic that buoyancy effect causes Raw, ym represents the impact that fast turbulence pulsation can be pressed to expand to total dissipative shock wave, σ_k、σ_εIt is Turbulent Kinetic and its dissipative shock wave respectively Turbulent prandtl number.

4. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that institute Stating gas-particle two-phase flow movable model is Lagrangian stochastic particle model trajectory.

5. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that institute Stating gas phase turbulance combustion model is Hybrid analysis probability density function (pdf) model.

6. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that institute State the diffusion kinetics control including for fugitive constituent precipitation two-equation model, burning for coke of pulverized coal particle combustion model Combustion model.

7. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that institute Stating radiative heat transfer model is p-1 radiation patterns.

8. power boiler breeze combustion characteristics according to claim 1 generates the analogy method of numerical value it is characterised in that institute Stating nox generation model is pdf transport equation model.

9. a kind of power boiler breeze combustion characteristics generates the analog of numerical value it is characterised in that including:

Division unit, for carrying out multiblock technique division to preset boiler-burner model；

Analogue unit, for by basal conservation eqution, turbulent flow model, gas-particle two-phase flow movable model, gas phase turbulance Combustion model, pulverized coal particle combustion model, radiative heat transfer model and nox generation model are to the described preset boiler after stress and strain model The combustion process of the boiler coal-ash in burner model is simulated；

Determining unit, for determining the airflow field in described boiler-burner, described boiler-burner according to the result after simulation Smoke components distribution in interior temperature field, described boiler-burner, motion rail in described boiler-burner for the pulverized coal particle Mark, described boiler-burner Wall Radiation hot-fluid, boiler combustion efficiency and no_xDischarge capacity.

10. power boiler breeze combustion characteristics according to claim 9 generate numerical value analog it is characterised in that Also include:

Unit reprocessed by model, for simplifying to the described preset boiler-burner model after stress and strain model, and gets Design parameter after setting and working condition.