A numerical method for predicting NOx emissions from coal-ammonia blended combustion x emissions

Abstract

The application discloses a numerical calculation method for predicting NO x emissions in coal-ammonia blended combustion, and belongs to the technical field of coal-ammonia co-combustion numerical simulation. The method steps include: establishing gas-solid two-phase control equations, characterizing differential diffusion through non-unity Lewis number; discretizing the model and coupling the velocity-pressure field through the PIMPLE algorithm; carrying out CP-DNS simulation considering differential diffusion and NO x generation; calculating the general mixing fraction suitable for coal-ammonia blended combustion based on the CP-DNS data set, incorporating differential diffusion in the small flame model through non-unity constant Lewis number, generating a small flame lookup table containing differential diffusion, evaluating the model performance through prior analysis, combining reaction path, chemical time scale and budget analysis, and clearly defining the NO x generation mechanism to support emission prediction. The application can fully capture the influence of differential diffusion and multi-dimensional diffusion on NO x generation.

Description

Technical Field

[0001] This invention belongs to the field of numerical simulation technology for coal-ammonia co-combustion, specifically relating to a method for predicting NO in coal-ammonia co-combustion. x Numerical calculation methods for emissions. Background Technology

[0002] Coal-ammonia co-combustion, as a low-carbon emission technology, has significant advantages in reducing CO2 emissions, while the addition of ammonia alters NO2 emissions. x Generation mechanism, prediction of NO x Emissions are crucial for the industrial application of this technology. In coal-ammonia co-combustion systems, large molecules in volatiles coexist with small molecules such as NH3 and H2, resulting in significant differential diffusion effects that influence combustion characteristics and NO emissions. x The generation process has a significant impact. Carrier phase direct numerical simulation (CP-DNS) can resolve all spatiotemporal scales of turbulence and chemistry, providing high-precision datasets for combustion mechanism research; the small flame model can consider detailed chemical reaction mechanisms while ensuring computational efficiency, making it an effective tool for studying complex combustion processes. The differential diffusion model applicable to gas-phase combustion is extended to the coal-ammonia blended combustion system, and the differential diffusion is characterized by non-unity Lewis numbers in the gas-solid two-phase governing equations and the small flame model, respectively, to investigate NO. x Generation mechanism, improving model building assumptions and enhancing NO x The reliability of emissions prediction is of great importance. Summary of the Invention

[0003] This invention provides a method for predicting NO in coal-ammonia blended combustion. x Numerical methods for calculating emissions extend the differential diffusion model of gas-phase combustion to coal-ammonia blended combustion systems. Differential diffusion is characterized in both the gas-solid two-phase governing equations and the small flame model. Based on a high-precision dataset obtained from CP-DNS, multi-dimensional analysis is used to explore NO emissions. x Generation mechanism and model construction hypotheses.

[0004] The technical solution of the present invention is as follows:

[0005] A method for predicting NO in coal-ammonia blended combustion x The numerical calculation method for emissions includes the following steps:

[0006] Step 1: Characterize the differential diffusion effect in the gas-solid two-phase model using the non-uniform Lewis number and establish the governing equations;

[0007] Step 2: Couple the pressure-velocity field and conduct direct numerical simulation to obtain combustion dataset, thereby generating CP-DNS data; CP-DNS represents direct numerical simulation of the carrier phase.

[0008] Step 3, based on the CP-DNS data, the general mixing fraction Z suitable for coal-ammonia mixed combustion is calculated by the Bilger formula, based on the general mixing fraction Z, the non-premixed small flame equation is solved by the non-unity constant Lewis number, the small flame lookup table containing difference diffusion is generated, and the small flame model is constructed;

[0009] Step 4, by prior analysis, the prediction values of the small flame model containing difference diffusion are compared with the CP-DNS data, and the prediction effects of temperature and components are evaluated; combined with reaction path analysis, chemical time scale analysis and budget analysis, the NO x generation mechanism is clarified, and the NO x emission prediction of coal-ammonia mixed combustion is realized.

[0010] The beneficial effects of the present application are as follows:

[0011] 1. The gas-phase difference diffusion model is extended to the coal-ammonia mixed combustion system, and the difference diffusion effect is captured by the non-unity Lewis number in the gas-solid two-phase control equation and the small flame model, which is suitable for the complex characteristics of coal-ammonia mixed combustion;

[0012] 2. Based on the high-precision CP-DNS data set considering difference diffusion, the general mixing fraction suitable for coal-ammonia mixed combustion is calculated by the Bilger formula, and the small flame model containing difference diffusion is constructed, which provides a solid foundation for the NO x generation mechanism research and emission prediction;

[0013] 3. Combined with prior analysis, reaction path analysis, chemical time scale analysis and budget analysis, the NO x generation mechanism and model construction hypothesis are systematically explored, which takes into account the calculation accuracy and efficiency, and provides a reliable numerical tool for the optimization design of coal-ammonia co-combustion burner and the NO x emission prediction. BRIEF DESCRIPTION OF DRAWINGS

[0014] Figure 1 A flowchart of a numerical calculation method for predicting NO x emission in coal-ammonia mixed combustion according to the present application;

[0015] Figure 2 A schematic diagram of the coal-ammonia mixed combustion CP-DNS calculation domain according to the present application;

[0016] Figure 3 A prior analysis result graph of the small flame model containing difference diffusion according to the present application-temperature;

[0017] Figure 4 A prior analysis result graph of the small flame model containing difference diffusion according to the present application-H2O mass fraction;

[0018] Figure 5 Prior analysis result graph of the small flame model with differential diffusion of the present application-C6H6O2 mass fraction;

[0019] Figure 6 Prior analysis result graph of the small flame model with differential diffusion of the present application-H2 mass fraction;

[0020] Figure 7 Prior analysis result graph of the small flame model with differential diffusion of the present application-NO mass fraction;

[0021] Figure 8 Prior analysis result graph of the small flame model with differential diffusion of the present application-HCN mass fraction;

[0022] Figure 9 Budget analysis graph of the Lagrange-like transient equation of the present application;

[0023] Figure 10 Budget analysis graph of the generalized small flame equation of the present application. DETAILED DESCRIPTION

[0024] The present application will be described in detail below with reference to the accompanying drawings and specific embodiments. However, the following embodiments are only used to explain the present application, and the protection scope of the present application should include the entire content of the claims, and through the description of the following embodiments, those skilled in the art can fully realize the entire content of the claims of the present application.

[0025] Embodiment:

[0026] This embodiment takes coal-ammonia mixed combustion in a turbulent mixing layer as the research object, and details the implementation process of the present application (see Figure 1 ):

[0027] Step 1, establish relevant mathematical models:

[0028] 1. Gas-solid two-phase control equation: the gas phase scalar transport equation adopts the following form, which covers the species mass fraction, total enthalpy and various mixing fractions; the differential diffusion effect in the control equation is characterized by a non-unity constant Lewis number; the mature model is used to calculate the volatile analysis and the coke surface reaction rate, and the corresponding gas-solid two-phase coupling source term is included in the control equation. The general form of the gas phase scalar transport equation of the gas-solid two-phase control equation is:

[0029] ,

[0030] wherein, is a general scalar corresponding to the species mass fraction , total enthalpy , or ; For gas phase density, Euler time, For directional fluid velocity, The index of three-dimensional spatial coordinates representing the Cartesian coordinate system ( Corresponding to direction), For gas phase thermal conductivity, The non-unity constant Lewis number characterizes the differential diffusion effect in the governing equation; For gas-specific heat capacity, This term represents the chemical reaction rate; only the equation for the mass fraction of species contains this term. It is a gas-solid two-phase coupled source term, originating from volatile matter and coke surface reaction.

[0031] 2. Extension of the Differential Diffusion Model: The differential diffusion model applicable to gas-phase combustion is extended to coal-ammonia blended combustion systems to explore the role of differential diffusion in coal-ammonia blended combustion and its effect on NO. x The impact on predictive performance; this extended model is also based on the non-unity Lewis number principle, ensuring consistency with the diffusion characterization logic of the gas-solid two-phase control equation.

[0032] Step 2, Numerical Discretization and Iterative Solution:

[0033] 1. Mesh Generation: The finite volume method was used to discretize the governing equations. Spatial derivatives were processed using a cubic discretization scheme, and temporal derivatives were discretized using an implicit scheme. A cuboid computational domain was used, with flow, transverse, and spanwise dimensions of 59.7 mm × 59.7 mm × 14.9 mm, a mesh resolution of 384 × 384 × 96, and a uniform mesh spacing of 155.5 μm. This ensured that the multi-scale characteristics of turbulent vortices and the flame front were fully analyzed. The boundary conditions of the computational domain were set to a coal / ammonia / air mixture in the upper laminar flow. , , Temperature 600K, speed 15m / s These represent the mass fractions of NH3, O2, and N2, respectively; the lower laminar flow is hot air at a temperature of 1300 K and a velocity of -15 m / s; the coal powder particles have a diameter of 25 μm and a number density of 1 × 10⁻⁶. 11 The particle count is approximately 1 m³, with initial velocity and temperature consistent with the gas phase. The x and z directions have periodic boundaries, while the y direction has a zero-gradient boundary. Initial turbulent disturbances are generated using a digital filtering method, and their intensity is... Length scale . Figure 2 This is a schematic diagram of the CP-DNS calculation domain setting for coal-ammonia blended combustion according to the present invention.

[0034] 2. Discretization and solution: spatial derivatives are discretized with a third-order scheme, and time integration is performed with an implicit scheme. Nonlinear algebraic equations are generated and solved iteratively at each time step to ensure stability.

[0035] Step 3, CP-DNS solution and data set acquisition:

[0036] 1. Numerical solution: a low-Mach number finite volume solver based on OpenFOAM is used to solve the Navier-Stokes equations. The pressure implicit operator splitting and pressure coupled equation hybrid algorithm (PIMPLE) algorithm is used to couple the velocity-pressure field, and OpenSMOKE++ is used to solve the detailed chemical reaction mechanism. The effects of differential diffusion and NO x The generated CP-DNS simulation is run until the simulation reaches a steady state.

[0037] 2. Data set acquisition: 3D data sets containing temperature, mass fraction of each component, and mixture fraction are extracted.

[0038] Step 4, small flame model construction and mechanism analysis:

[0039] 1. General mixture fraction calculation and lookup table generation: based on the CP-DNS data set, the general mixture fraction Z suitable for coal-ammonia blended combustion is calculated by the Bilger formula.

[0040] The general mixture fraction Z suitable for coal-ammonia blended combustion is calculated based on the Bilger formula:

[0041] ,

[0042] ,

[0043] where and are the values on the fuel side and the oxidant side, respectively, is the mixture fraction of elements C, H, and O, is the molecular weight of elements C, H, and O. The mass fraction of the fuel side component is calculated by the triple mixture fraction model:

[0044] ,

[0045] ,

[0046] ,

[0047] are the gas mass from ammonia, volatile, and char tail gas, respectively. , , mass fraction of component in ammonia stream, volatile, and char tail gas, respectively , , mixing fraction of ammonia, mixing fraction of volatile, and mixing fraction of char tail gas, respectively is the char tail gas ratio, quantifying the mixing relationship between char tail gas and volatile is the coal-ammonia mixing ratio, distinguishing the dominant position of coal and ammonia combustion. The general mixing fraction The specific calculation formula is:

[0048] .

[0049] wherein, represents the amount of component in the combustion system, is the mass fraction of component , is the number of atoms of C, H, and O in component , respectively, represents the molecular weight of component .

[0050] Prior analysis focuses on evaluating the prediction effect of temperature, main components and NO x by comparing the predicted values of small flame model with different diffusivity and CP-DNS data; the small flame lookup table FLT is generated by FlameMaster software, is the progress variable, represents the mass fraction of CO2, H2O, and H2, respectively; is the normalized total enthalpy, is the total gas-phase enthalpy, and are the minimum and maximum total enthalpy of the total gas-phase enthalpy , respectively. By incorporating the non-unity constant Lewis number into the small flame model through the above manifold coordinates, the non-premixed small flame equation is solved to generate the small flame lookup table with different diffusivity:

[0051] The non-premixed small flame equation includes species transport equation and energy equation, which are:

[0052] ,

[0053] ,

[0054] wherein, is the gas-phase density, is the scalar dissipation rate, represents the component ​Lewis number, the reaction rate of component , the heat release rate, the specific heat capacity of gas phase, the specific heat capacity of component , the average molecular weight of mixture, the gas phase temperature. The calculation is based on a detailed chemical reaction mechanism containing 129 species and 1664 elementary reactions.

[0055] 2. Priori analysis: extract manifold coordinates from CP-DNS data, compare the predicted values of small flame model with difference diffusion and CP-DNS data (see Figure 3 , Figure 4 , Figure 5 , Figure 6 , Figure 7 and Figure 8 , FLT-DD represents flame surface lookup table with difference diffusion), the results show that the prediction effect of temperature and main component is good, and the prediction of NO mass fraction exists deviation.

[0056] 3. Reaction path analysis: quantify the NO x generation path by calculating the nitrogen atom flux ratio:

[0057] ,

[0058] where, is used to define the reaction path flux ratio of nitrogen atom from component P to component Q. Where, is the total number of elementary reactions, is the total number of species; is the flame index, , represent the mass fraction of C 16 H 10 , CO, NH3 respectively, represents the gradient symbol; is the mass fraction of O2; is the flux of nitrogen atom between species P and Q in reaction , is the nitrogen atom flux between NH3 and component . The net consumption rate of NH3 in premixed combustion zone ( ) and non-premixed combustion zone ( ) is 0.121 kg / (m s) and 0.145 kg / (m s) respectively, and the corresponding net NO flux is 0.001 kg / (m s) and 0.016 kg / (m s), premixed combustion mode has no significant effect on overall NO production.

[0059] 4. Chemical time scale analysis: By estimation, the chemical characteristic time scale of species ; is the gas phase density; is the mass fraction of species , is the net reaction rate of species , is a small positive number to verify the applicability of the steady small flame assumption; NO, NO2, etc. NO x species chemical time scale is much smaller than CO2, H2O, etc. main components, verify the applicability of the steady small flame assumption.

[0060] 5. Budget analysis: Budget analysis includes Lagrangian transient equation and generalized small flame equation, the relationship between the Lagrangian-like transient term and Eulerian transient term for general scalar is:

[0061] ,

[0062] In the formula, represents the Eulerian transient term, is the Lagrangian-like transient term, is the transient contribution of the mixing fraction field, is the convection term of the small flame coordinate along the mixing fraction isoparametric surface; wherein, is the Euler time, is the Lagrangian-like time, is the gas phase density, is the fluid velocity vector, is the isoparametric surface unit normal vector, represents the gradient symbol.

[0063] The budget equation of the generalized small flame equation is:

[0064] ,

[0065] is the normal diffusion term along the mixture fraction gradient direction, is the multi-dimensional diffusion term along the isoparametric surface, which is ignored in the one-dimensional small flame equation, is the differential diffusion term, is the source term, is the correction term; the Lagrangian transient term is mainly balanced by the transient contribution of the mixing fraction field according to the Eulerian transient term budget analysis (see Figure 9 ), although the Lagrangian transient term is crucial for NO, it can be partially considered in the a priori analysis because the popular coordinate originates from the DNS data, and this term has been included in the DNS calculation process; the generalized small flamelet equation budget analysis shows that the one-dimensional small flamelet model neglecting the multi-dimensional diffusion effect is the main factor leading to the deviation of NO prediction in the non-premixed combustion zone (see Figure 10 , the gray area in the figure represents the reaction zone).

[0066] This embodiment shows that the present application can effectively explore the NO x formation mechanism and model construction hypothesis by representing the differential diffusion effect in the gas-solid two-phase control equation and the small flamelet model, combined with the CP-DNS high-precision data set analysis, to provide reliable numerical tools and theoretical support for the prediction of NO x emission and the optimization design of the combustion system of coal-ammonia mixed combustion.

[0067] The above description is merely a specific implementation of the present application, enabling those skilled in the art to understand or implement the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined in the present application can be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to these embodiments shown in the present application, but will conform to the widest scope consistent with the principles and novel features of the present application.

Claims

1. A numerical calculation method for predicting NOx emission in coal-ammonia co- combustion, characterized by, x The method comprises the following steps: ​ Step 1, difference diffusion effect is characterized in gas-solid two-phase model by non-unity Lewis number, and control equation is established; Step 2, pressure-velocity field is coupled, direct numerical simulation is carried out to obtain combustion data set, and CP-DNS data are generated; CP-DNS represents carrier phase direct numerical simulation; Step 3, based on CP-DNS data, general mixing fraction Z suitable for coal-ammonia mixed combustion is calculated by Bilger formula, and based on the general mixing fraction Z, non-premixed small flame equation is solved by non-unity constant Lewis number, small flame lookup table containing difference diffusion is generated, and a small flame model is constructed; Step 4, by priori analysis, compare the small flame model prediction value with the CP-DNS data, evaluate the prediction effect of temperature and components; combine reaction path analysis, chemical time scale analysis and budget analysis, and clarify the NO x generation mechanism, realize the NO x emission prediction of coal-ammonia mixed combustion.

2. A method for predicting NO in coal-ammonia blended combustion according to claim 1. x The numerical calculation method for emissions is characterized by, In step 1, the general form of the gas phase scalar transport equation of the control equation is: ， wherein, is the mass fraction of the species i, is the total enthalpy, , or ; is the gas phase density, is the Euler time, is the directional fluid velocity, represents the three-dimensional spatial coordinate index of the Cartesian coordinate system, respectively correspond to directions, is the gas phase thermal conductivity, is the non-unity constant Lewis number, representing the differential diffusion effect in the governing equations; is the gas phase specific heat capacity, is the chemical reaction rate, is the gas-solid two-phase coupling source term.

3. A method for predicting NO in coal-ammonia blended combustion according to claim 2. x The numerical calculation method for emissions is characterized by, In step 3, the general mixing fraction Z suitable for coal-ammonia mixed combustion is calculated based on the Bilger formula: ， ， wherein , , is the mixing fraction of the elements C, H, O, , , is the molecular weight of the elements C, H, O; and are the values of the fuel side and the oxidizer side value, respectively; Fuel side component mass fraction Calculated by ternary mixing fraction model: ， ， ， wherein, , , are the mass fractions of the components in the ammonia stream, the mass fractions of the components in the volatiles, and the mass fractions of the components in the coke tail gas, respectively; , , are the gas masses of the ammonia, the volatiles, and the coke tail gas, respectively, , , are the mixing fractions of the ammonia, the volatiles, and the coke tail gas, respectively; is the coke tail gas proportion; is the coal-ammonia mixing ratio.​​​ 4. A method for predicting NO in coal-ammonia blended combustion according to claim 3. x The numerical calculation method for emissions is characterized by, Difference diffusion is included in the small flame model by non-unity constant Lewis number, and a small flame lookup table is generated by solving the non-premixed small flame equation, the non-premixed small flame equation includes species transport equation and energy equation, respectively: ， ， where is the gas phase density, is the mass fraction of component , is the scalar dissipation rate, represents the Lewis number of component , is the reaction rate of component , is the heat release rate, is the gas phase temperature, represents the number of components in the combustion system; , is the gas phase specific heat capacity, is the specific heat capacity of component , is the average molecular weight of the mixture, Z is the mixture fraction, and was calculated using a detailed chemical reaction mechanism containing 129 species and 1664 elementary reactions.

5. A method for predicting NO in coal-ammonia blended combustion according to claim 4. x The numerical calculation method for emissions is characterized by, The prior analysis in step 4 compares the predicted values ​​of the small flame model with differential diffusion with CP-DNS data, focusing on assessing temperature, composition, and NO. x The predictive effect; For progress variables, , , These represent the mass fractions of CO2, H2O, and H2, respectively. For normalized total enthalpy, For total enthalpy in the gas phase, and Total enthalpy of the gas phase The minimum and maximum total enthalpy.

6. A method for predicting NO in coal-ammonia blended combustion according to claim 1. x The numerical calculation method for emissions is characterized by, In step 1, the control equation is discretized by finite volume method, the spatial derivative is processed by cubic discrete format, and the time derivative is discretized by implicit format; the grid division adopts a cuboid calculation domain, the flow direction, transverse direction and spanwise size are 59.7mm*59.7mm*14.9mm respectively, the grid resolution is 384*384*96, and the uniform grid spacing is 155.5μm.

7. A method for predicting NO in coal-ammonia blended combustion according to claim 1. x The numerical calculation method for emissions is characterized by, Boundary conditions in step 2 are set as: the upper stream is coal / ammonia / air mixture, , , , temperature 600 K, velocity 15 m / s, , , represent the mass fraction of NH3, O2, N2, respectively; the lower stream is hot air, temperature 1300 K, velocity -15 m / s; the diameter of coal powder particle is 25 μm, the number density is 1 x 10 11 particles / m³, the initial velocity and temperature are consistent with the gas phase; the x and z directions are periodic boundary, and the y direction is zero gradient boundary.

8. A method for predicting NO in coal-ammonia blended combustion according to claim 7. x The numerical calculation method for emissions is characterized by, Reaction path analysis in Step 4 quantifies NO by calculating the ratio of nitrogen atom flux x Path of generation: ， wherein, for defining the reaction path flux ratio of nitrogen atoms from species P to species Q; wherein, is the total number of elementary reactions, is the total number of species; is the flame index, represents the gradient sign, , , , represent the mass fractions of C 16 H 10 , CO, NH3, respectively; is the mass fraction of O2; is the flux of nitrogen atoms between species P and species Q in the reaction , is the flux of nitrogen atoms between NH3 and species .

9. A method for predicting NO in coal-ammonia blended combustion according to claim 1. x The numerical calculation method for emissions is characterized by, The chemical timescale in step 4 is estimated by , The chemical characteristic timescale for the species ; The gas phase density; The mass fraction of the species , The net reaction rate of the species , is positive, verifying the applicability of the steady-state small flame assumption.

10. A method for predicting NO in coal-ammonia blended combustion according to claim 9. x The numerical calculation method for emissions is characterized by, The budget analysis in Step 4 employs a Lagrangian transient equation and a generalized small flame budget equation, a relationship between the Lagrangian transient term and the Eulerian transient term with respect to a generic scalar is: ， wherein represents the Euler transient term, is a Lagrangian-like transient term, is a mixing fraction field, is a small flame coordinate along the mixing fraction contour surface; wherein, is the Euler time, is a Lagrangian-like time, is the gas phase density, is the fluid velocity vector, is is the unit normal vector of the contour surface, represents the gradient sign; The budget equation of the generalized small flame equation is: ， where, is the normal diffusion term along the mixture fraction gradient direction, is the multi-dimensional diffusion term along the iso-surface, which is neglected in one-dimensional small flamelet equations, is the differential diffusion term, is the source term, is the correction term.

Images