Numerical simulation method and device for combustion thermoacoustic instability and pollutant emission of hydrogen-doped natural gas

Abstract

The invention provides a numerical simulation method and device for combustion thermoacoustic instability and pollutant emission of hydrogen-doped natural gas, and belongs to the technical field of combustion of hydrogen-doped natural gas. The method provided by the invention comprises the following steps: establishing a gas turbine combustion cylinder model; writing a three-dimensional LBM program of the combustion cylinder model of the gas turbine; a temperature additional item of the simulated combustion process is calculated, a collision equation of the three-dimensional LBM program is corrected based on the temperature additional item, and the temperature additional item is used for representing the heat-fluid-solid coupling state in the combustion process; calculating a turbulence viscosity coefficient according to the combustion cylinder model of the gas turbine, and calculating the real-time turbulence viscosity of the combustion process based on the turbulence viscosity coefficient by the corrected three-dimensional LBM program; the three-dimensional LBM program simulates the combustion process in the combustion cylinder of the gas turbine based on the input combustion parameters and the calculated real-time turbulence viscosity. The numerical simulation method and device for thermoacoustic instability and pollutant emission of combustion of hydrogen-doped natural gas are used for improving the simulation accuracy of the combustion process of the hydrogen-doped natural gas.

Description

Technical Field

[0001] This application relates to the field of hydrogen-blended natural gas combustion technology, and in particular to a numerical simulation method and apparatus for thermoacoustic instability and pollutant emissions during hydrogen-blended natural gas combustion. Background Technology

[0002] As the global energy structure shifts towards cleaner and lower-carbon fuels, hydrogen-blended natural gas, as a new type of fuel that combines economic efficiency and environmental friendliness, is experiencing continuous growth in demand for applications such as gas turbine power generation and industrial heating. By blending a certain proportion of hydrogen into natural gas, it can effectively reduce carbon emissions during combustion while improving fuel combustion efficiency, aligning with the core trend of current sustainable energy development. Research on related combustion characteristics has become one of the key directions in the energy sector.

[0003] Currently, the industry's analysis of hydrogen-blended natural gas combustion processes relies primarily on experimental testing and numerical simulation. Numerical simulation, due to its advantages of low cost, short cycle time, and ability to cover complex operating conditions, has become the main research approach. Existing simulation methods are mostly based on the traditional computational fluid dynamics framework. By establishing combustion reaction kinetic models and combining them with turbulent flow equations, they calculate the flow field, temperature field, and component distribution of hydrogen-blended natural gas within the combustion device, providing fundamental data support for combustion system design. However, existing numerical simulation techniques still have significant limitations. On the one hand, most methods struggle to accurately characterize the thermo-fluid-structure interaction effect during combustion, neglecting the influence of energy transfer between the wall and the fluid on the combustion state, leading to significant deviations between the temperature field and actual operating conditions. On the other hand, the calculation of turbulent viscosity often relies on empirical formulas, failing to adjust parameters in real time according to the flow field characteristics of the combustion region. This makes it difficult to accurately reflect the effects of turbulent motion on combustion stability and pollutant generation, ultimately resulting in low accuracy in thermoacoustic instability prediction and insufficient reliability of pollutant emission simulation results, failing to meet the needs of refined design for hydrogen-blended natural gas combustion systems. Summary of the Invention

[0004] In view of this, this application provides a numerical simulation method and apparatus for the thermoacoustic instability and pollutant emissions of hydrogen-blended natural gas combustion, so as to improve the accuracy of the simulation of the hydrogen-blended natural gas combustion process.

[0005] Specifically, this application is implemented through the following technical solution:

[0006] The first aspect of this application provides a numerical simulation method for the thermoacoustic instability of hydrogen-blended natural gas combustion and pollutant emissions, the method comprising:

[0007] Establish a gas turbine combustion chamber model;

[0008] A three-dimensional LBM program for the gas turbine combustor model was developed, which was used to simulate the combustion process of multi-component fluids in hydrogen-blended natural gas.

[0009] The temperature addition term of the combustion process is calculated in real time, and the collision equation of the three-dimensional LBM program is corrected based on the temperature addition term. The temperature addition term is used to characterize the thermal-fluid-structure coupling state during the combustion process.

[0010] The turbulent viscosity coefficient is calculated based on the gas turbine combustor model, and the corrected three-dimensional LBM program calculates the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient.

[0011] The three-dimensional LBM program simulates the combustion process inside the gas turbine combustor based on real-time input combustion parameters and calculated real-time turbulent viscosity.

[0012] A second aspect of this application provides a numerical simulation device for the thermoacoustic instability of hydrogen-blended natural gas combustion and pollutant emissions. The device includes a construction module, a calculation module, and a simulation module; wherein...

[0013] The construction module is used to build a gas turbine combustion chamber model;

[0014] The construction module is also used to write a three-dimensional LBM program for the gas turbine combustor model, which is used to simulate the combustion process of multi-component fluids in hydrogen-blended natural gas.

[0015] The calculation module is used to calculate the temperature addition term of the simulated combustion process in real time, and correct the collision equation of the three-dimensional LBM program based on the temperature addition term. The temperature addition term is used to characterize the thermal-fluid-structure coupling state during the combustion process.

[0016] The calculation module is also used to calculate the turbulent viscosity coefficient based on the gas turbine combustor model, and the corrected three-dimensional LBM program calculates the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient.

[0017] The simulation module is used by the three-dimensional LBM program to simulate the combustion process inside the gas turbine combustor based on the real-time input combustion parameters and the calculated real-time turbulent viscosity.

[0018] The numerical simulation method and apparatus for thermoacoustic instability and pollutant emissions during the combustion of hydrogen-blended natural gas provided in this application improve the accuracy of numerical simulation of the combustion process of hydrogen-blended natural gas in a gas turbine combustor by real-time calculation of temperature add-ons to incorporate the effects of thermal-fluid-structure interaction and dynamic adaptation of turbulent viscosity coefficients to synchronize with actual turbulent conditions. This provides reliable support for thermoacoustic instability analysis and pollutant emission prediction. Specifically, by establishing a gas turbine combustor model that fits the actual application scenario, a precise geometric and physical basis is provided for the numerical simulation of the hydrogen-blended natural gas combustion process, avoiding simulation errors caused by model-to-actual deviations. A three-dimensional LBM program is written based on the gas turbine combustor model to specifically simulate the combustion process of multi-component fluids in hydrogen-blended natural gas. Compared with traditional simulation methods, this method is more suitable for the complex characteristics of multi-component mixing and reaction of hydrogen-blended fuels, laying the program framework foundation for subsequent accurate simulation. By real-time calculation of temperature add-ons during the combustion process and using this to correct the three-dimensional LBM program, the effects of thermal-fluid-structure interaction during the combustion process can be effectively incorporated. This compensates for the problem of traditional simulation neglecting energy transfer at the wall and fluid interface, leading to temperature field calculation distortion, and further improves the accuracy of the simulation. The simulation is accurate; simultaneously, the turbulent viscosity coefficient is calculated based on the gas turbine combustor model, allowing the modified 3D LBM program to accurately obtain the real-time turbulent viscosity of the combustion process. This ensures that the depiction of turbulent motion during the simulation is synchronized with the actual combustion conditions, avoiding simulation deviations in combustion state caused by fixed turbulent parameters. Ultimately, the 3D LBM program simulates the combustion process based on the real-time input combustion parameters and the calculated real-time turbulent viscosity, achieving precise control over the entire process from model construction and program optimization to dynamic parameter adaptation. This significantly improves the accuracy of simulating the combustion process of hydrogen-blended natural gas in the gas turbine combustor, providing reliable numerical simulation support for subsequent in-depth analysis of combustion thermoacoustic instability phenomena and accurate prediction of pollutant emissions, and assisting in the optimized design and stable operation of hydrogen-blended natural gas combustion systems. Attached Figure Description

[0019] Figure 1 A flowchart of an embodiment of the numerical simulation method for thermoacoustic instability of hydrogen-blended natural gas combustion and pollutant emissions provided in this application;

[0020] Figure 2 This is a schematic diagram of the second embodiment of the numerical simulation device for thermal and acoustic instability of hydrogen-blended natural gas combustion and pollutant emissions provided in this application. Detailed Implementation

[0021] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings represent the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application.

[0022] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used herein are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0023] It should be understood that although the terms first, second, third, etc., may be used in this application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish information of the same type from one another. For example, without departing from the scope of this application, first information may also be referred to as second information, and similarly, second information may also be referred to as first information. Depending on the context, the word "if" as used herein may be interpreted as "when," "when," or "in response to determination."

[0024] The following specific embodiments are given to illustrate the technical solution of this application in detail.

[0025] Figure 1 This is a flowchart of Example 1 of the numerical simulation method for the thermoacoustic instability of hydrogen-blended natural gas combustion and pollutant emissions provided in this application. Please refer to... Figure 1 The method provided in this embodiment may include:

[0026] S101. Establish a gas turbine combustion chamber model.

[0027] Specifically, the gas turbine combustor is the component in a gas turbine responsible for mixing fuel (such as natural gas or hydrogen-blended natural gas) with air and performing efficient combustion, converting chemical energy into thermal energy. Specific structural data of the combustor is obtained through surveying or design drawings, clarifying geometric information such as the inlet and outlet positions, internal flow guidance structure, and cavity dimensions. Then, the physical properties of the wall material, such as thermal conductivity, density, and specific heat capacity, are determined. Simultaneously, boundary conditions for the combustor are set, such as whether the walls are insulated and the pressure or flow velocity conditions at the inlet and outlet. Finally, this information is integrated to construct a model that accurately reflects the actual physical environment of the combustor.

[0028] Furthermore, the specific implementation steps for establishing the gas turbine combustor model include:

[0029] (1) Obtain the structural parameters of the combustion chamber of the gas turbine in industrial operation;

[0030] Specifically, the structural components of an industrial gas turbine combustor include the combustion chamber, fuel nozzle layout, swirler structure, intake channel, exhaust port size, and wall thickness. Professional surveying tools (such as laser scanners and coordinate measuring machines) are used to conduct on-site measurements of the operating industrial-grade combustor to obtain specific dimensional data for each structural component, such as chamber diameter, length, nozzle orifice diameter and spacing, and swirler blade angle and number. Simultaneously, it is necessary to collect physical parameters of the combustor wall material (such as thermal conductivity and density) and morphological data of the internal flow channels (such as the presence of flow guide structures and flow channel curvature) to provide complete structural and physical information support for subsequent modeling.

[0031] (2) Based on the geometric similarity criterion, the scaling factor is determined using the actual size of the industrial gas turbine combustor as the benchmark;

[0032] Specifically, the modeling objectives and experimental constraints are determined, the maximum accommodating size of the model is determined, and the scaling factor is determined based on the maximum accommodating size and the actual size. For example, if the diameter of an industrial combustion chamber is 1 meter, but the laboratory bench can only accommodate a model with a diameter of 0.2 meters, then the initial scaling factor is determined to be 1:5. The initial scaling factor needs to be verified to confirm whether the scaled dimensions of each component (such as nozzles and cyclones) still meet the requirements for machining accuracy and experimental measurement. If any component is too small to be machined after scaling, the scaling factor needs to be adjusted appropriately to ensure that the scaled model is both geometrically similar and practically operable.

[0033] (3) Using three-dimensional modeling software, a gas turbine combustion chamber model is established based on the structural parameters and the scaling factor.

[0034] Specifically, the collected structural data and determined scaling factors are converted into a visualized 3D model that can be used for subsequent simulations or experiments. In practice, first, a suitable 3D modeling software (such as SolidWorks or ANSYS DesignModeler) is selected, and the acquired industrial combustor structural parameters are imported. Then, based on the scaling factors, all structural parameters are proportionally converted to obtain the dimensional data of each component in the model. Further, in the software, according to the actual combustor assembly relationship, 3D models of components such as the cavity, nozzle, cyclone separator, intake channel, and exhaust channel are constructed sequentially, ensuring that the relative positions and connection methods of each component are consistent with the prototype. Finally, the model is optimized in detail, such as repairing geometric defects that occur during modeling (such as gaps between surfaces) and dividing a reasonable mesh (if used for numerical simulation), generating a complete 3D model of the gas turbine combustor that meets geometric similarity requirements.

[0035] By collecting structural parameters of industrial combustion chambers and determining the scaling ratio using geometric similarity criteria, model distortion caused by missing parameters or imbalanced scaling is avoided. This ensures that the established model accurately reflects the internal flow channel morphology, component layout, and relative dimensional relationships of the actual combustion chamber, providing a reliable geometric basis for subsequent combustion process simulations. The scaling factor is determined by fully considering the limitations of laboratory bench space, processing precision, and computational resources. This avoids the problem of models being too large to fit the experimental environment and also prevents the situation where models are too small, making it difficult to process or measure key components. At the same time, a reasonable scaling model can reduce the computational load of numerical simulations, improving computational efficiency and reducing resource consumption while ensuring simulation accuracy. The established 3D model can be directly used for subsequent numerical simulations (such as LBM program development and combustion thermoacoustic instability simulation) and experimental verification. Furthermore, because the model follows geometric similarity criteria, subsequent research results can be extrapolated to industrial-grade combustion chambers through similarity principles. This provides transferable experimental and simulation basis for the study of combustion characteristics and thermoacoustic instability control of industrial gas turbines burning hydrogen-blended natural gas, reducing the cost and risk of conducting experiments directly on industrial equipment.

[0036] S102. Write a three-dimensional LBM program for the gas turbine combustor model, which is used to simulate the combustion process of multi-component fluids in hydrogen-blended natural gas.

[0037] Specifically, the 3D LBM program refers to a numerical calculation program developed based on the Lattice Boltzmann Modeling (LBM) method and adapted to the 3D model of a gas turbine combustor. The LBM method is a numerical simulation technique between the molecular scale and the macroscopic scale. By simulating the collision and migration process of particles on a discrete lattice, it indirectly restores the laws of mass conservation, momentum conservation and energy conservation of macroscopic fluid motion. It is naturally adapted to the simulation of multi-component flow, complex boundaries (such as the combustor wall) and unsteady dynamic processes (such as thermoacoustic pulsation).

[0038] Furthermore, the geometric parameters of the combustion chamber model are transformed into an LBM computational mesh, incorporating the physical properties (such as density and specific heat capacity), diffusion coefficients, and combustion reaction kinetics of each component in the hydrogen-blended natural gas. This allows for the construction of LBM governing equations for multi-component fluid flow and combustion reaction. The 3D LBM program leverages LBM's advantages in handling multi-component flow and complex boundary problems to accurately capture the microscopic flow, component mixing, and chemical reaction details during the combustion of hydrogen-blended natural gas, providing an effective computational tool for dynamic simulation of the combustion process. The 3D LBM program constructs a computational mesh based on a 3D scaled model of the gas turbine combustion chamber, accurately reproducing the geometric features such as the flow channel morphology and swirler structure within the combustion chamber. Simultaneously, it incorporates the physical properties (density, diffusion coefficient) and combustion reaction kinetics (such as those based on detailed mechanisms in GRI 3.0) of the multi-component fluids (methane, hydrogen, oxygen, etc.) in the hydrogen-blended natural gas, achieving coupled simulation of multi-component mixing and combustion reaction.

[0039] Furthermore, the specific implementation steps for writing the 3D LBM program for the gas turbine combustor model include:

[0040] (1) Based on the three-dimensional features and flow field spatial distribution of the gas turbine combustor model, the basic framework of the three-dimensional LBM program is built based on the D3Q19 spatial discretization format;

[0041] Specifically, the three-dimensional coordinate information of the gas turbine combustor model is analyzed to determine the boundary range and mesh generation method of the flow field computational domain, ensuring that the mesh can fit the complex structure inside the combustor (such as the flow field abrupt change region near the cyclone separator and nozzle); the migration path and collision rules of particles are defined based on the D3Q19 spatial discretization format, and the basic lattice Boltzmann equations are established, covering the discretized expression of mass conservation and momentum conservation; the migration and collision of particles on the mesh nodes and the extraction of macroscopic physical quantities (density, velocity) are realized through programming, forming a basic computational framework that can support the integration of subsequent modules.

[0042] Furthermore, the D3Q19 spatial discretization scheme is a discrete velocity model used in the Lattice Boltzmann Method (LBM) for three-dimensional flow field simulation, widely applied in the numerical simulation of complex three-dimensional physical processes such as multi-component flow and combustion reactions. It transforms continuous fluid motion into particle migration and collision processes on a discrete lattice by defining 19 discrete particle velocity directions in three-dimensional space, thereby achieving numerical solutions for macroscopic flow fields. D3 represents three-dimensional space with x, y, and z coordinate directions, and Q19 indicates the setting of 19 discrete velocity vectors in this three-dimensional space. These velocity vectors, centered on a node of the computational grid, cover the possible directions of particle motion: including one zero-velocity direction, indicating that particles do not migrate at that node, and 18 non-zero velocity directions distributed along the positive and negative directions of the three-dimensional coordinate axes and the four spatial diagonals, forming a symmetrical velocity distribution pattern. This discretization method ensures the ability to capture anisotropic flows in three-dimensional space while achieving a balance between computational accuracy and efficiency by reasonably simplifying the number of velocity directions.

[0043] In practical applications, the D3Q19 spatial discretization scheme describes the particle density and momentum characteristics in each velocity direction by defining a corresponding particle distribution function. Particles migrate along 19 preset directions within a discrete time step and then collide at grid nodes. The collision process follows a simplified form of the Boltzmann equation, and the discretization of mass, momentum, and energy conservation is achieved by adjusting the distribution function. This scheme is particularly suitable for simulating complex three-dimensional flow fields within gas turbine combustors. It can accurately characterize the three-dimensional flow characteristics induced by structures such as channel bends and cyclones, and by coupling modules for multi-component transport and combustion reactions, it supports the coupled simulation of multi-physics fields during the combustion of hydrogen-blended natural gas, providing a reliable numerical solution framework for the dynamic evolution of flow field, temperature field, and component concentration field.

[0044] (2) Based on the composition of hydrogen-blended natural gas in the gas turbine combustor model, a multi-component transport and combustion reaction module is added to the basic framework. The multi-component transport and combustion reaction module is used to simulate the coupling process of multi-component mixing and combustion reaction.

[0045] Specifically, for the composition of hydrogen-blended natural gas (such as methane, hydrogen, oxygen, and combustion products), a multi-component transport and combustion reaction module is developed and integrated within the basic framework to achieve coupled simulation of mixing and reaction. The physical properties of each component in the hydrogen-blended natural gas are determined, and a component transport equation is established based on a multi-component LBM model. This equation describes the diffusion and convection processes of different components in the flow field. Combustion reaction kinetics are introduced, embedding the reaction source term into the component transport equation, and real-time calculation of the reaction rate is achieved through programming. A coupling algorithm links component transport and the combustion reaction process, enabling the module to dynamically output the concentration distribution of each component and the reaction exothermic rate, providing input for temperature field calculations.

[0046] Furthermore, the core components of hydrogen-blended natural gas are defined, typically including fuel components (methane, hydrogen), oxidant (oxygen), combustion products (carbon dioxide, water vapor), and inert gases (nitrogen, from air), determining the proportion range of each component in the initial mixed gas. Physical properties of each component are obtained from a literature database, including molecular mass, diffusion coefficient, and specific heat capacity. These physical property parameters are converted into dimensionless forms suitable for LBM calculations, ensuring consistency of parameters across different components in numerical calculations. For each component, a corresponding particle distribution function is defined within the D3Q19 discrete velocity framework. This particle distribution function characterizes the particle density of the component in a specific velocity direction. Based on the continuous medium assumption of multi-component LBM, transport equations are established for each component. These equations describe the diffusion motion of components due to concentration gradients. By summing the particle distribution functions in each velocity direction, the concentration and diffusion flux of each component are obtained. Based on these concentrations and diffusion fluxes, a quantitative description of component diffusion and convection processes in the flow field is achieved.

[0047] Furthermore, based on the combustion characteristics of hydrogen-blended natural gas, suitable reaction kinetics are selected. For each component, its reaction source term is determined by the elementary reaction participating in the reaction. The reaction source term is transformed into a correction term of the particle distribution function layer and embedded into the collision term in the component transport equation to realize the dynamic influence of chemical reaction on component concentration changes.

[0048] (3) Based on the wall structure and heat transfer characteristics of the gas turbine combustion tube model, a thermal-fluid-structure interaction calculation interface is generated in the basic frame. The thermal-fluid-structure interaction calculation interface is used to simulate the energy transfer at the interface between the wall and the fluid.

[0049] Specifically, by combining the wall structure (such as wall thickness and material properties) and heat transfer characteristics (such as thermal conductivity and radiation characteristics) of the gas turbine combustor model, a thermal-fluid-structure interaction (TFS) calculation interface is developed within the basic framework to simulate the energy exchange between the fluid and the wall. The inputs to the TFS calculation interface are the wall structure parameters of the gas turbine combustor model and real-time flow field data. The wall structure parameters include wall thickness, material properties, and the initial temperature of the wall mesh nodes. The real-time flow field data is determined by the deviation between the near-wall fluid temperature, fluid velocity, energy probability density distribution function, and equilibrium state output from the current iteration step of the 3D LBM program. The output is a temperature addition term under the TFS interaction, which exists in the form of an energy source term. This temperature addition term quantifies the instantaneous energy exchange intensity between the wall and the fluid. The TFS calculation interface directly passes the output temperature addition term to the collision equations of the 3D LBM program, embedding it into the energy evolution module to correct the iterative updates of the fluid energy field. This ensures real-time coupling between the flow field temperature distribution and the wall heat transfer characteristics, improving the accuracy of the TFS simulation.

[0050] Furthermore, the geometric boundary conditions of the combustion chamber wall are defined (such as the coordinates and types of the wall mesh nodes), and a heat transfer model between the fluid and the wall (including convective heat transfer and heat conduction) is established based on the thermo-fluid-structure interaction theory. An interface module is developed through programming to acquire the temperature field on the fluid side and the wall temperature in real time, calculate the interfacial heat flux density, and convert it into an energy source term to feed back into the energy equation of the LBM. Simultaneously, the interface also needs to consider the thermal response of the wall material (such as the change in wall temperature over time) to ensure the bidirectional and dynamic nature of the energy transfer simulation. The instantaneous temperature of the wall mesh nodes is introduced as a dynamic variable, with its initial value set by the initial temperature of the wall material. The wall surface temperature is determined and iteratively updated over time steps, directly characterizing the thermal response state of the wall material. Based on the laws of heat conduction and energy conservation, a calculation equation for the wall temperature change over time is established. At each time step, the updated instantaneous wall temperature is calculated using the equation, reflecting the wall's thermal response after receiving fluid energy. The instantaneous wall temperature is used as an input parameter and fed into the calculation model for the temperature supplementary term to obtain the corrected fluid energy source term. This corrected fluid energy source term is then output to the collision equation of the 3D LBM program, thus influencing the evolution of the flow field temperature. Changes in the flow field temperature will have a feedback effect on the calculation of the instantaneous wall temperature in the next time step. For details on the specific implementation process of the generation interface, please refer to the descriptions in relevant technologies; they will not be elaborated here.

[0051] (4) Based on the turbulent flow characteristics of the gas turbine combustor model, a turbulent viscosity calculation module is generated in the basic framework. The turbulent viscosity calculation module is used to simulate the energy transfer between the energy-efficient vortex and the dissipative vortex.

[0052] Specifically, based on the turbulent characteristics of the flow field inside the gas turbine combustor model (such as turbulence intensity and vortex scale distribution), a turbulent viscosity calculation module is developed in the basic framework to simulate the energy transfer between energy-efficient vortices and dissipative vortices in turbulence. By analyzing the generation mechanism of turbulence in the combustion chamber (such as wall shear and disturbances caused by abrupt changes in the flow channel), a suitable turbulence model (such as Large Eddy Simulation (LES)) is selected. Based on the model theory, the correlation between turbulent viscosity and flow field parameters (such as turbulent kinetic energy and strain rate) is derived. LBM describes the macroscopic flow field based on the evolution of microscopic particle distribution functions. The calculation of turbulent viscosity needs to be adapted to the discretization format of LBM (such as the particle velocity components and relaxation time parameters of the D3Q19 model). Even if a suitable turbulence model is selected, the macroscopic parameters such as turbulent kinetic energy and strain rate in the model need to be converted into microscopic quantities that can be calculated under the LBM framework (such as particle distribution function deviation and energy probability density function gradient), and the correlation between the two needs to be derived (such as binding the strain rate with the velocity gradient term in LBM). Only in this way can the turbulent viscosity be called in real time during the LBM collision-migration process, ensuring the compatibility between the turbulence model and numerical methods. By implementing real-time calculation logic through programming, the module needs to extract parameters such as velocity gradient and pulsation from the flow field and dynamically update the turbulent viscosity value. Finally, the calculation results are embedded into the momentum equation to correct the viscous force term of the fluid, so as to accurately reflect the influence of turbulent motion on momentum transport.

[0053] Furthermore, standard model coefficients are introduced into the derivation of the correlation between turbulent viscosity and flow field parameters based on model theory. The coefficients at this stage are preset values ​​of the standard equation and are used to build the basic calculation logic. Subsequently, the standard model coefficients can be updated in a targeted manner according to the differences in turbulent characteristics of different regions in the combustion chamber (such as the main combustion zone and the near-wall zone). For example, the coefficients in the main combustion zone are increased to enhance the turbulent diffusion effect. Finally, the calculation results are embedded into the momentum equation to correct the viscous force term of the fluid, so as to accurately reflect the influence of turbulent motion on momentum transfer.

[0054] (5) Adjust the parallel computing architecture of the three-dimensional LBM program according to the calculation scale and accuracy requirements of the gas turbine combustion tube model.

[0055] Specifically, based on the computational scale and accuracy requirements of the gas turbine combustor model, the parallel computing architecture of the 3D LBM program is optimized to improve computational efficiency. First, the computational load of the model (e.g., number of mesh nodes, number of iterations) is assessed to determine the parallelization strategy (e.g., MPI parallelism based on domain decomposition or GPU acceleration). Then, based on the spatial distribution characteristics of the flow field (e.g., the difference in computational intensity between the combustion core and peripheral regions), load-balanced sub-computation domains are divided to ensure a relatively balanced computational load across parallel processes. Finally, the data communication interface is optimized through programming to reduce data exchange latency between processes, and memory allocation and caching strategies are adjusted for high-precision computational needs (e.g., detailed simulation of chemical reactions) to improve program efficiency while ensuring computational accuracy.

[0056] By constructing a three-dimensional LBM program, the three-dimensional flow field characteristics inside the gas turbine combustor can be accurately reproduced, and the entire process of multi-component mixing and combustion reaction of hydrogen-blended natural gas can be realistically simulated. The thermal-fluid-structure interaction effect can be effectively incorporated to improve the realism of the temperature field simulation, and the turbulent energy transfer can be accurately characterized to optimize the simulation accuracy of combustion stability. At the same time, by adjusting the parallel computing architecture to balance the simulation efficiency and accuracy requirements, a high-precision dynamic simulation of multi-physics coupling in the combustion process of hydrogen-blended natural gas can be achieved, providing a high-fidelity numerical tool for studying combustion thermoacoustic instability and pollutant emissions.

[0057] Furthermore, the implementation steps for writing a 3D LBM program for a gas turbine combustor model also include:

[0058] (1) Determine the component concentrations of multiple target components of the reaction source item based on the chemical reaction mechanism of hydrogen-doped natural gas, and calculate the reaction rate of each elementary reaction in the hydrogen-doped natural gas;

[0059] Specifically, the core chemical reaction mechanism of hydrogen-blended natural gas combustion is clearly defined (such as the elementary reaction chain involving methane and hydrogen oxidation), and key target components participating in the reaction are selected (such as reactants H2, CH4, and O2, intermediate products OH and CH3, and products CO2 and H2O). The concentration of target components at each grid node in the flow field is obtained through the component transport module of the multi-component LBM model. Then, according to the Arrhenius rate formula of the elementary reaction, the reaction rate of each elementary reaction under the current concentration and temperature conditions is calculated. The reaction rate reflects how fast the reaction proceeds.

[0060] (2) Calculate the net generation rate of each target component based on the reaction rate and the component concentration;

[0061] Specifically, for each target component, all elementary reactions it participates in are statistically analyzed (including reactions in which it is consumed as a reactant and reactions in which it is generated as a product). Based on the stoichiometric coefficients, the rate of each elementary reaction is multiplied by the stoichiometric coefficient of that component in the reaction, where consumption is negative and generation is positive. The contributions of all related reactions are then summed to obtain the net generation rate of that component. A positive net generation rate indicates that the component is generated, and a negative rate indicates that it is consumed, directly reflecting the total effect of the chemical reaction on the change in component concentration.

[0062] (3) The net generation rate is converted into a reaction source term form suitable for the evolution of the LBM probability density function, and the reaction source term is embedded in the collision equation of the probability density function of each component to simulate the coupling between combustion reaction and flow field.

[0063] Specifically, the net generation rate of the target component is converted into a reaction source term at the probability density function level, ensuring that its unit is consistent with the time rate of change of the distribution function. Then, the reaction source term is embedded into the collision equation of the probability density function of each component as an additional correction term in the collision process, so that the distribution function can not only reflect the convection and diffusion of the flow field in the migration-collision iteration, but also respond to the increase or decrease of components caused by chemical reactions. Through this embedding, the dynamic coupling of component concentration changes in the combustion reaction and momentum and energy transfer in the flow field is realized, so that the simulation results simultaneously reflect the transport effect of the flow field on the reaction and the feedback effect of the reaction on the flow field.

[0064] This allows for precise quantification of the dynamic impact of chemical reactions on the concentrations of various components in hydrogen-blended natural gas, and deep coupling of these effects with the flow and diffusion processes of the flow field. This overcomes the limitations of traditional simulations that separate the calculation of reactions and flow fields, improves the realism of the simulation of the interaction between combustion reactions and flow fields, and lays a reliable foundation at the chemical reaction level for the subsequent accurate simulation of temperature fields, thermoacoustic instability, and pollutant emissions.

[0065] S103. The temperature addition term of the combustion process is calculated in real time, and the collision equation of the three-dimensional LBM program is corrected based on the temperature addition term. The temperature addition term is used to characterize the thermal-fluid-structure coupling state during the combustion process.

[0066] Specifically, the calculation of the temperature addition requires a comprehensive analysis of the heat exchange between the fluid and the combustion chamber wall during combustion, the influence of wall temperature changes on the thermal state of the fluid, and other thermo-fluid-structure interaction effects. By quantifying these effects, correction parameters characterizing the thermo-fluid-structure interaction state are obtained.

[0067] Furthermore, a heat transfer model between the fluid and the wall is established. Based on the real-time simulated flow field temperature and wall temperature, the heat flux density is calculated, and a temperature-related term is derived. This term is then incorporated into the collision equations of the 3D LBM program to correct its thermophysical calculation module. This approach compensates for the shortcomings of traditional simulations that neglect the thermal-fluid-structure interaction, making the 3D LBM program's simulation of the combustion temperature field more realistic, improving the accuracy of temperature field calculations, and enhancing the overall realism of the combustion process simulation.

[0068] Furthermore, the specific implementation steps for the temperature additional term in the real-time calculation simulation of the combustion process include:

[0069] (1) Determine the calculation boundary constraints of the temperature addition term based on the wall thermal boundary conditions of the gas turbine combustor model;

[0070] Specifically, the wall thermal boundary types of the combustion chamber model are analyzed, such as adiabatic walls with zero heat flux, isothermal walls with a fixed wall temperature, or convective heat transfer walls where heat exchange occurs between the wall and the external environment. For different types of boundaries, constraint rules for temperature additional terms are set. For example, for adiabatic walls, the temperature additional terms are constrained to prevent the generation of additional heat flux at the wall. For isothermal walls, the temperature additional terms are constrained to maintain the wall temperature at a preset value. For convective heat transfer walls, the external heat transfer coefficient is included in the boundary constraints. These constraints are transformed into mathematical conditions (such as the heat flux continuity equation at the wall and temperature gradient limits) as the computational boundary constraints for calculating temperature additional terms.

[0071] (2) Obtain the calculation parameters of the temperature additional term based on the instantaneous thermophysical parameters of the flow field during the simulated combustion process;

[0072] Specifically, during the combustion simulation in the 3D LBM program, instantaneous thermophysical parameters of each grid node in the flow field are collected in real time, including the fluid's temperature distribution, density, specific heat capacity, thermal conductivity, and the flow velocity at the fluid-wall interface, where the flow velocity represents the intensity of convective heat transfer. At the same time, the real-time temperature of the wall, the thermal conductivity of the wall material, and the thickness are obtained through the thermal-fluid-structure interaction calculation interface. These parameters are matched according to the spatial grid nodes to form a dynamic dataset for calculating additional temperature terms, ensuring that the parameters are updated in real time as the flow field evolves.

[0073] (3) Based on the boundary constraints and the calculation parameters, construct a real-time calculation model for the temperature additional term based on the apparent enthalpy method, and calculate the temperature additional term based on the real-time calculation model.

[0074] It should be noted that the temperature addition must satisfy two constraints simultaneously. The first is the energy conservation constraint, which means that the flow field energy equation after the temperature addition must satisfy the heat flux conservation at the fluid-wall interface, ensuring that the heat flux output from the fluid side is equal to the heat flux input from the wall side. The second is the boundary condition constraint, which must strictly follow the wall thermal boundary conditions. For adiabatic wall scenarios, the temperature addition must keep the interface heat flux at zero after it is applied. For isothermal wall scenarios, the temperature addition must compensate for the flow field state that deviates from the set temperature to ensure that the wall temperature is stable at the preset value.

[0075] Specifically, the relationship between fluid enthalpy and temperature is defined, and the temperature-related additional term is converted into an enthalpy-related additional term for calculation. The convective heat transfer flux on the fluid side is calculated based on the instantaneous thermophysical parameters of the flow field, while the heat conduction flux on the wall side is also calculated. Based on energy conservation constraints, the enthalpy-related additional term, i.e., the enthalpy correction corresponding to the temperature-related additional term, is derived to ensure that the additional term can compensate for the difference in heat flow between the fluid and the wall. The enthalpy correction is then adjusted in conjunction with boundary condition constraints, ultimately converting the enthalpy-related additional term into a temperature-related additional term for real-time calculation. Through this model, the temperature-related additional term can be dynamically updated with the flow field thermophysical parameters and the wall state, accurately reflecting the instantaneous effects of thermo-fluid-solid coupling.

[0076] Furthermore, based on the thermophysical properties of the fluid, enthalpy is defined as the correlation between specific heat capacity and temperature, so that changes in enthalpy can directly correspond to changes in temperature. The temperature addition is indirectly obtained by calculating the correction of enthalpy. Based on the real-time thermophysical parameters of the flow field, such as fluid temperature and flow velocity, the convective heat transfer intensity between the fluid and the wall is determined, and the heat flow transferred from the fluid side to the wall is obtained. Furthermore, combined with the thermal conductivity characteristics of the wall material, such as thermal conductivity and thickness, and the temperature difference between the wall and the external environment, the heat flow conducted by the wall itself is calculated, that is, the heat that the wall dissipates to the external environment or absorbs from the outside. Based on the energy conservation constraint, the difference between the heat flow on the fluid side and the heat flow on the wall side is used as the enthalpy correction. This difference is compensated by adjusting the enthalpy correction. If the heat transferred by the fluid is more than the heat conducted by the wall, the fluid energy needs to be reduced through the enthalpy addition term to maintain balance, and vice versa. The enthalpy addition term is adjusted according to the boundary condition constraints, and then the adjusted enthalpy addition term is converted into a temperature addition term according to the correlation between enthalpy and temperature. Finally, the temperature correction parameter that can reflect the heat-fluid-solid coupling state in real time is obtained.

[0077] By clearly defining boundary constraints, dynamically acquiring parameters, and constructing a real-time calculation model, the temperature addition term can accurately quantify the dynamic heat exchange between the fluid and the wall during combustion. This overcomes the limitations of simplified handling of thermal-fluid-structure interaction effects in traditional simulations, improves the realism of temperature field simulation, and provides more reliable thermal environment data support for subsequent thermoacoustic instability analysis.

[0078] S104. Calculate the turbulent viscosity coefficient based on the gas turbine combustor model, and then calculate the real-time turbulent viscosity of the combustion process based on the corrected three-dimensional LBM program.

[0079] Specifically, the turbulent viscosity coefficient is a calculated term in the turbulence model. The turbulent viscosity coefficient changes with the structural parameters of the combustor. The turbulent viscosity coefficient is calculated based on the gas turbine combustor model, and real-time turbulent viscosity is obtained. Combining the internal structural features and flow field characteristics of the combustor model, a turbulent viscosity coefficient calculation method matching the combustor structure and flow state is derived based on turbulence model theory. Within the code framework, parameter calling interfaces related to combustor structural features, such as channel curvature and obstacle distribution, and flow field characteristics, such as velocity gradient and disturbance intensity, are embedded. The turbulent viscosity coefficient calculation method derived from turbulence model theory is transformed into modular code. This modular code can include the mapping relationship between structural parameters and flow field parameters, and the code implementation of the dynamic solution formula for the coefficient. After code integration, when running the 3D LBM program, the turbulent viscosity calculation module automatically reads the structural parameters and real-time flow field data of the combustor model, solves for the turbulent viscosity coefficient according to the preset calculation method, and then drives the dynamic output of real-time turbulent viscosity, realizing the program's adaptive calculation of combustor structural changes and flow field evolution. The revised 3D LBM program utilizes the turbulent viscosity coefficient, combined with real-time simulated flow field parameters (such as velocity fluctuations), to dynamically calculate the real-time turbulent viscosity during combustion. This overcomes the limitations of traditional methods that rely on fixed empirical values ​​to calculate turbulent viscosity, enabling real-time responses to flow field changes during combustion. It more accurately characterizes the impact of turbulent motion on combustion reaction and heat transfer, thus improving the accuracy of combustion stability simulation.

[0080] Furthermore, the specific steps for calculating the turbulent viscosity coefficient based on the gas turbine combustor model include:

[0081] (1) Extract the model parameters of the gas turbine combustor model and collect flow field data; wherein, the flow field data includes the particle distribution function, equilibrium particle distribution function, temperature gradient and energy transfer characterization vector of each grid point;

[0082] Specifically, structural parameters (such as channel diameter, wall roughness, and internal component dimensions) and physical property parameters (such as thermal conductivity of wall material and fluid reference density) of the combustion chamber model are extracted. Through the real-time monitoring module of the 3D LBM program, microscopic and macroscopic data of each grid point in the flow field are collected, including the distribution function of particles in 19 discrete velocity directions to reflect the particle motion state, the equilibrium particle distribution function in the corresponding velocity direction to represent the particle distribution under ideal conditions, the gradient change of flow field temperature in space to reflect temperature non-uniformity, and the energy transfer characterization vector to describe the direction and intensity of energy transfer in the flow field. These data are stored by grid node index to form a dynamically updated dataset.

[0083] (2) Calculate the flow field fluctuation value based on the particle distribution function and the equilibrium particle distribution function;

[0084] Specifically, for each grid point in the flow field, the actual particle distribution function is compared with the equilibrium particle distribution function in each velocity direction, and the deviation between the two is calculated. By statistically integrating the deviations in all velocity directions (such as sum of squares or modulus calculation), the flow field fluctuation value of that grid point is obtained. The larger the flow field fluctuation value, the stronger the randomness of particle motion in the flow field, the more intense the turbulent fluctuation, and directly reflects the strength of the pulsating kinetic energy in the turbulent motion.

[0085] (3) Calculate the turbulence enhancement term induced by thermal convection based on the temperature gradient and the energy transfer characterization vector;

[0086] Specifically, the spatial distribution characteristics of the temperature gradient are analyzed to determine the intensity of natural convection caused by temperature differences; the larger the temperature gradient, the more significant the natural convection. The energy transfer characterization vector is used to clarify the energy transfer path and direction in the flow field and determine whether the energy is transformed into turbulent vortex motion. Based on the coupling relationship between the temperature gradient and the energy transfer characterization vector, the turbulence enhancement term induced by thermal convection is calculated. The turbulence enhancement term is used to describe the additional turbulent energy generated during thermal convection and reflects the role of heat transfer in turbulence development under non-isothermal conditions.

[0087] (4) Calculate the turbulent viscosity coefficient based on the flow field fluctuation value and the turbulence enhancement term.

[0088] Specifically, the flow field fluctuation value is used as a quantitative indicator of the ground-state energy of turbulence, reflecting the inherent fluctuation intensity of turbulent motion; the turbulence enhancement term induced by thermal convection is used as an additional energy indicator, reflecting the strengthening effect of thermal convection on turbulence; based on the theory of turbulent energy transfer, the flow field fluctuation value and the turbulence enhancement term are integrated according to weights, and the weights of the flow field fluctuation value and the turbulence enhancement term are dynamically adjusted according to the turbulent characteristics of the flow field in the combustion tube, to obtain a parameter that comprehensively reflects the intensity of turbulent motion; then, combined with the strain rate characteristics of the flow field extracted from the velocity field, this parameter is converted into a turbulent viscosity coefficient. The larger the final output coefficient value, the stronger the turbulence's resistance to momentum transfer, and the more accurately it can reflect the influence of turbulence on the combustion process.

[0089] This approach incorporates both the inherent pulsations of the flow field and the turbulence enhancement effect induced by thermal convection, overcoming the limitations of fixed empirical formulas. It achieves real-time matching between the turbulent viscosity coefficient and the dynamic characteristics of the flow field within the combustion chamber, improving the simulation accuracy of turbulent vortex energy transfer and momentum dissipation processes. This provides more reliable turbulent parameter support for the analysis of combustion stability and thermoacoustic instability phenomena in hydrogen-blended natural gas.

[0090] Furthermore, the specific implementation steps of the revised three-dimensional LBM program for calculating the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient include:

[0091] (1) Determine the variables for real-time turbulent viscosity calculation according to the turbulent viscosity calculation formula in the three-dimensional LBM program, wherein the variables include at least model constants;

[0092] Specifically, based on the above description, the turbulent viscosity calculation module pre-embeds the code implementation of the turbulent viscosity calculation formula. This means that the flow field parameters, combustion chamber structure-related parameters, and model constants involved in the formula are transformed into variables and calculation logic that the module can call, making the formula the core calculation unit of the module. During subsequent execution of the 3D LBM program, real-time parameters are first extracted from the flow field data pool of the LBM main program, and then the embedded turbulent viscosity calculation formula is called for numerical solution, directly outputting the turbulent viscosity coefficient. Ultimately, this provides data support for the calculation of real-time turbulent viscosity, achieving deep integration between the formula and the program module and automated solution.

[0093] Furthermore, the calculation formulas for turbulent viscosity in the 3D LBM program (such as the correlation formula based on turbulent kinetic energy and dissipation rate) are analyzed. All variables in the formula are sorted out. In addition to dynamic parameters such as flow field fluctuation values ​​and thermal convection enhancement terms, model constants in the formula are identified, such as the turbulent Prandtl number and proportionality coefficient. These constants are fixed parameters set based on turbulence theory or experience. These variables are classified and organized to clarify which are dynamic quantities that change with the flow field in real time and which are model constants that need to be set or adjusted in advance, so as to ensure the integrity of variable input in subsequent calculations.

[0094] (2) Adjust the model constants according to the flow field characteristics of different combustion zones in the gas turbine combustor;

[0095] Specifically, the combustion chamber is divided into different combustion zones (such as the fuel-air mixing zone, the main combustion zone, and the burnout zone). The turbulence characteristics of each zone are clarified through flow field analysis. For example, the mixing zone is dominated by shear turbulence, while the main combustion zone has strong thermal convection turbulence due to the heat released during combustion. Based on the turbulence generation mechanism of each zone (such as shear action and thermal driving action), combined with experimental data or theoretical analysis, the model constants are adjusted differentially. For example, the weight of model constants related to thermal convection is increased in the main combustion zone, and the influence of constants related to shear pulsation is strengthened in the mixing zone, so that the adjusted model constants can accurately reflect the dominant factors of regional turbulence.

[0096] Furthermore, by experimentally measuring turbulent characteristic parameters (such as turbulence intensity, vortex scale, heat flux density, etc.) in different regions within the combustion chamber, or by deriving the dominant mechanisms of turbulence generation and dissipation in each region based on turbulence theory: For the fuel-air mixing zone, experimental data typically show that turbulence in this region is mainly driven by shear forces (such as strong shear induced by the velocity gradient at the cyclone outlet). Combining shear turbulence theory, the model constants related to velocity fluctuations can be increased (such as the shear term coefficient) to highlight the contribution of shearing to turbulent viscosity; in the main combustion zone, due to... Combustion releases heat, generating intense thermal convection. Experimentally measured temperature gradients and heat flux fluctuations are significant. Based on thermal convection turbulence theory, it is necessary to increase the model constants related to thermal convection (e.g., by increasing the heat flux-turbulence coupling coefficient) to make the turbulent viscosity more sensitive to the thermally driven vortex enhancement. In the burnout zone, the flow field tends to stabilize, turbulence intensity weakens, and dissipation becomes dominant. Referring to turbulence dissipation theory and combining it with experimentally measured low-fluctuation data, the overall amplitude of the model constants can be reduced, while the weights of individual sub-terms can be decreased to match the characteristics of turbulent energy decay in this region. By calibrating through experimental data and guided by theoretical mechanisms, the model constants for different regions can be accurately adapted to their dominant turbulence factors, achieving regional differentiation and physical realism in turbulent viscosity calculations.

[0097] (3) In each time step iteration of the three-dimensional LBM program, the real-time turbulent viscosity of each grid point is directly calculated by combining the adjusted model constants.

[0098] Specifically, in each time step iteration of the 3D LBM program, the real-time flow field parameters (such as flow field fluctuation values, thermal convection enhancement terms, velocity gradients, etc.) of each grid point are called; according to the combustion zone to which the grid point belongs, the corresponding adjusted model constants are matched; the real-time flow field parameters and model constants are substituted into the turbulent viscosity calculation formula, and the real-time turbulent viscosity value is obtained by solving grid by grid; the calculation results are fed back to the momentum equation of LBM to correct the fluid viscous force term in that time step, ensuring that the turbulent viscosity can respond to the instantaneous changes in the flow field in real time.

[0099] Furthermore, before starting the turbulent viscosity calculation, a pre-simulation without turbulent viscosity solution is performed using a 3D LBM program to obtain the basic flow field parameters of the global grid points. Based on the basic flow field parameters, the degree of flow field variation between each grid point is quantified. Representative grid points are selected based on the analysis results of the degree of flow field variation. During the iteration process, calculations are performed on the representative grid points.

[0100] Specifically, the similarity of flow field characteristics is determined by calculating the Euclidean distance or rate of change difference of parameters of adjacent grid points. For continuous grid regions with highly consistent flow field trends, i.e., parameter differences less than a set threshold, they are merged into a single computational unit. A representative point can be selected from the origin of this region, or a virtual representative point can be re-determined after weighted averaging of the flow field parameters of the merged region. Only grid points with drastic flow field changes and unique characteristics are retained as independent computational units. Real-time flow field parameter calls, model constant matching, and turbulent viscosity calculations are performed on the selected representative grid points. The turbulent viscosity values ​​of the remaining grid points are quickly obtained by combining the calculation results of the representative grid points with spatial mapping algorithms (such as linear interpolation and regional feature extrapolation). Ultimately, this achieves full-domain grid coverage with less computation, improving iteration efficiency while ensuring computational accuracy in key areas.

[0101] By calculating turbulent viscosity in real time, the model constants are preserved in accordance with turbulence theory. Regional adjustments are made to adapt to the turbulent characteristics of different combustion regions. At the same time, time-step iteration is combined to achieve dynamic synchronization with the flow field evolution. This overcomes the shortcomings of traditional fixed-constant calculation of turbulent viscosity, which makes it difficult to take into account regional differences and instantaneous changes. It significantly improves the spatiotemporal accuracy of turbulent viscosity calculation and provides more reliable parameter support for the simulation of the coupling effect of turbulence and combustion in hydrogen-blended natural gas combustion.

[0102] S105. The three-dimensional LBM program simulates the combustion process inside the gas turbine combustor based on the real-time input combustion parameters and the calculated real-time turbulent viscosity.

[0103] Specifically, the 3D LBM program receives real-time input combustion parameters (such as hydrogen blending ratio, inlet flow rate, initial temperature, etc.), and combines them with the previously calculated real-time turbulent viscosity. By solving the discrete lattice Boltzmann equations, it dynamically updates physical quantities such as the flow field, temperature field, and component concentration field within the combustion chamber. It iterative calculations are performed according to the time step, continuously updating the physical parameters of each grid node to simulate the dynamic evolution of the combustion process. This enables dynamic and accurate simulation of the entire combustion process of hydrogen-blended natural gas within the combustion chamber, effectively capturing thermoacoustic instability phenomena and the generation and emission patterns of pollutants during combustion, providing reliable numerical data for analyzing combustion characteristics and optimizing combustion system design.

[0104] Furthermore, the specific implementation steps include:

[0105] (1) Read the real-time input combustion parameters, convert the combustion parameters into the initial conditions of the particle distribution function of the inlet boundary, load the mesh parameters and wall thermal boundary conditions of the gas turbine combustion tube model, and complete the simulation initialization;

[0106] Specifically, the combustion parameters (such as hydrogen blending ratio, inlet flow rate, initial temperature and pressure) are read in real time. Based on the particle distribution function rules of the multi-component LBM model, the combustion parameters are transformed into the initial values ​​of the particle distribution function of each grid point at the inlet boundary of the combustion chamber, including the distribution functions corresponding to the initial density and velocity of each component. At the same time, the grid parameters (such as grid size, node coordinates, and computational domain range) and wall thermal boundary conditions (such as adiabatic, isothermal, or convective heat transfer type) of the gas turbine combustion chamber model are loaded to clarify the reflection or rebound rules of particles at the wall. The parameter initialization and boundary setting before simulation are completed to ensure that the calculation starting point conforms to the actual combustion conditions.

[0107] (2) The temperature addition term is embedded as a source term into the energy evolution module of the LBM collision equation to correct the spatial distribution of the temperature field;

[0108] Specifically, real-time temperature additions are obtained from the thermal-fluid-structure interaction (TFI) calculation interface. These temperature additions reflect the energy exchange between the fluid and the wall and are embedded as energy source terms in the energy evolution module of the LBM collision equation. During particle collisions at each grid point, the temperature additions adjust the enthalpy and temperature of the fluid by correcting the equilibrium deviation of the energy distribution function. This ensures that the spatial distribution of the temperature field not only reflects the heat release from combustion and convection in the flow field but also responds to the heat absorption or dissipation from the wall, thereby correcting the temperature field deviation caused by neglecting TFI and improving the realism of the temperature distribution.

[0109] Furthermore, the real-time generated temperature addition term is obtained from the thermal-fluid-structure interaction calculation interface. Based on the correlation between enthalpy and temperature, the energy correction amount corresponding to the temperature addition term is converted into an enthalpy source term, ensuring that its unit is consistent with the time rate of change of the energy distribution function. The module responsible for energy evolution in the LBM collision equation is located. This module updates the enthalpy and temperature of the flow field through the collision process of the energy distribution function. The converted enthalpy source term is embedded into the energy balance relationship of the collision equation as an additional correction term for the energy distribution function. During the particle collision stage of each grid node, the influence of the enthalpy source term is superimposed when the energy distribution function evolves towards the equilibrium state. If the wall transfers heat to the fluid, the enthalpy source term is positive, pushing the energy distribution function to shift towards a high enthalpy state, corresponding to an increase in temperature. If the fluid dissipates heat to the wall, the enthalpy source term is negative, causing the energy distribution function to adjust towards a low enthalpy state, corresponding to a decrease in temperature. For grid nodes near the wall, the enthalpy source term is boundary-corrected in combination with the wall thermal boundary conditions (such as adiabatic, isothermal) to ensure that the energy transfer at the wall meets the preset constraints.

[0110] (3) The real-time turbulent viscosity is embedded into the viscosity term of the LBM collision equation to correct the momentum exchange during the particle collision process;

[0111] Specifically, the real-time turbulent viscosity value output by the turbulent viscosity calculation module is called and converted into a relaxation time correction under the LBM framework. Turbulent viscosity is positively correlated with relaxation time and is embedded in the viscosity term of the collision equation. During the particle collision stage, the corrected relaxation time will adjust the rate at which the particle distribution function evolves towards the equilibrium state, so that the momentum exchange intensity between particles matches the real-time turbulent state. That is, the higher the turbulent viscosity, the more intense the momentum exchange, thereby accurately characterizing the dissipation and transfer of momentum by turbulent vortices and correcting the flow simulation deviation caused by the traditional fixed viscosity coefficient.

[0112] Furthermore, after obtaining the real-time turbulent viscosity values ​​of each grid point from the turbulent viscosity calculation module, based on the correlation between the viscosity term and relaxation time in LBM, the larger the turbulent viscosity, the longer the relaxation time for particles to evolve to equilibrium. The turbulent viscosity is then converted into a corresponding relaxation time correction. This conversion requires combining the physical properties of the fluid (such as reference density and lattice sound velocity) to ensure that the unit of the correction is consistent with the time scale of the LBM collision equation, forming a viscosity correction parameter that can be directly embedded; the viscosity term module of the LBM collision equation is then located. In the LBM collision equations, the viscosity term affects momentum exchange by controlling the rate at which the particle distribution function evolves towards equilibrium. The core of this is the relaxation time parameter in the BGK collision model. By identifying the core code segment responsible for calculating momentum exchange, the role of relaxation time in the collision rules is clarified. Specifically, relaxation time determines the intensity of momentum transfer during particle collisions by adjusting the rate of decay of the deviation between the actual distribution function and the equilibrium distribution function. The modified relaxation time correction replaces the fixed relaxation time parameter in the original collision equations, allowing the relaxation time at each grid point to dynamically change with real-time turbulent viscosity. In regions with high turbulent viscosity (such as the main combustion zone), the relaxation time is extended, the particle distribution function evolves more slowly towards equilibrium, and momentum exchange during collisions is more intense, simulating the strong momentum dissipation effect of turbulent vortices. In regions with low turbulent viscosity (such as the burnout zone), the relaxation time is shortened, momentum exchange is smoother, matching the stable physical state of the flow field. Meanwhile, in the collision calculation, the equilibrium deviation attenuation of the distribution function is recalculated based on the corrected relaxation time to ensure that the momentum exchange correction is accurately synchronized with the real-time turbulent state; finally, the boundary rules for the wall and special regions are adapted. For grid nodes near the combustion chamber wall, the relaxation time correction needs to be locally adjusted in combination with the wall boundary conditions (such as no-slip boundaries). For example, due to the enhanced viscosity effect at the wall, the relaxation time correction corresponding to the turbulent viscosity needs to be appropriately amplified to avoid momentum transfer distortion; for regions with abrupt changes in the flow field (such as the hydrocyclone outlet), the relaxation time difference between adjacent grid points needs to be smoothed through interpolation algorithms to prevent numerical oscillations caused by sudden changes in turbulent viscosity.

[0113] (4) Calculate the combustion reaction rate based on real-time combustion parameters;

[0114] Specifically, based on the real-time combustion parameters at the current time step (such as the concentration of each component and the temperature field distribution), combined with the chemical reaction mechanism of hydrogen-blended natural gas, the reaction rate calculation module is invoked to calculate the rate constant of each elementary reaction using the Arrhenius formula. Combined with the reactant concentration, the propulsion rate of a single reaction is calculated. Then, based on the stoichiometric coefficients, the net generation rate (reaction rate) of each component is obtained, which reflects the intensity of the combustion reaction in real time (such as a high reaction rate in the main combustion zone and a low rate in the burnout zone), providing a kinetic basis for subsequent component transport and energy release.

[0115] (5) Combine the modified viscosity term and the temperature additional term to perform the particle collision and migration process and complete the simulation of the combustion process.

[0116] Specifically, within each time step, the particle distribution function collision rules of each grid point are first updated based on the viscous term with real-time turbulent viscosity correction and the additional temperature term with thermal-fluid-structure interaction effect, combined with the combustion reaction rate. The particle migration process is then executed, causing the particles to move to adjacent grid nodes along the D3Q19 discrete velocity direction. After the migration is completed, collision calculations are performed again, and the flow field velocity, pressure, component concentration, and temperature field are updated through the evolution of the distribution function. The above iterations are repeated until the preset simulation duration is completed, and finally, the dynamic evolution results of the flow field, temperature field, and component concentration field inside the combustion chamber are output, realizing the simulation of the entire process of hydrogen-blended natural gas combustion.

[0117] The simulation process achieves real-time input of combustion parameters, deep coupling of thermo-fluid-structure interaction, turbulent dynamics and chemical reaction kinetics. The particle evolution at each time step can respond to the instantaneous changes in flow field, temperature, reaction and boundary conditions, overcoming the limitations of physical field separation calculation in traditional simulations. It significantly improves the dynamics and realism of the simulation of hydrogen-blended natural gas combustion process, and provides high-precision numerical simulation results for revealing the thermoacoustic instability mechanism and optimizing combustion system design.

[0118] Furthermore, the method provided in this embodiment also includes:

[0119] (1) Extract thermoacoustic instability characteristic parameters and pollutant concentrations from the simulation results;

[0120] Specifically, the extraction dimensions of thermoacoustic instability characteristic parameters are determined, including dynamic pressure parameters, such as the pressure pulsation amplitude and resonant frequency at each grid point, obtained by Fourier transform of pressure time history data; heat release parameters, such as heat release rate pulsation value and phase difference between heat release and pressure, calculated based on the instantaneous change of combustion reaction heat release rate; and flow field correlation parameters, such as vortex pulsation intensity and flame surface vibration frequency, extracted from the evolution data of velocity field and component concentration field. For pollutant concentration, the spatial concentration distribution and time average concentration of characteristic pollutants such as NOx and CO in combustion products in different combustion regions (such as main combustion zone and outlet region) are extracted. The spatial concentration distribution and time average concentration can be obtained by statistically analyzing pollutant concentration data at multiple time steps. The extracted parameters are classified and organized according to time series or spatial grid nodes to form a structured dataset (such as pressure pulsation spectrum map and pollutant concentration cloud map).

[0121] (2) Establish a pollutant mapping relationship based on the thermoacoustic instability characteristic parameters and pollutant concentration, and optimize the combustion system based on the pollutant mapping relationship.

[0122] Specifically, correlation analysis is performed on the extracted thermoacoustic instability parameters and pollutant concentration data. For example, the correlation trend between pressure pulsation amplitude and NOx concentration is determined. Generally, when thermoacoustic instability intensifies, the number of local high-temperature zones increases, NOx generation rises, and the relationship between heat release phase difference and CO concentration is examined. Phase imbalance may lead to incomplete combustion and increased CO concentration. A quantitative pollutant mapping model is established through data fitting or machine learning methods (such as regression analysis and neural networks) to clarify the degree of influence of changes in thermoacoustic instability parameters on pollutant concentration. Based on the mapping relationship, optimization targets for the combustion system are identified. If it is found that thermoacoustic instability and NOx concentration increase synchronously under high swirling intensity, the optimization direction focuses on adjusting the swirling structure. Specific optimization schemes are proposed, such as adjusting the spatial layout of fuel nozzles to reduce pressure pulsation amplitude or optimizing the secondary air ratio to improve combustion uniformity. At the same time, the effectiveness of the optimization schemes is verified by three-dimensional LBM simulation to ensure that pollutant emissions are reduced while suppressing thermoacoustic instability, thereby achieving a comprehensive improvement in the performance of the combustion system.

[0123] This approach not only quantifies the intrinsic relationship between thermoacoustic instability and pollutant emissions, avoiding the blind adjustment of single parameters in traditional optimization, but also accurately identifies weak points in the combustion system based on the mapping relationship, enabling the proposal of targeted optimization solutions. Simultaneously, simulation verification ensures the optimization effect, ultimately achieving synergistic optimization of combustion system thermoacoustic stability and low-pollution emissions, providing a scientific basis for parameter adjustment in the actual operation of gas turbines burning hydrogen-blended natural gas.

[0124] The numerical simulation method for thermoacoustic instability and pollutant emissions during combustion of hydrogen-blended natural gas provided in this embodiment establishes a gas turbine combustor model that closely matches industrial realities using geometric similarity criteria. This accurately recreates key structures such as internal flow channels and swirlers within the combustor, avoiding deviations between traditional simplified models and actual operating conditions, and providing reliable geometric and physical support for subsequent simulations. The three-dimensional LBM program, built based on the D3Q19 format, innovatively integrates multi-component transport and combustion reaction modules, dynamically characterizing the mixing and reaction coupling processes of components such as methane and hydrogen in hydrogen-blended natural gas. Simultaneously, it incorporates energy transfer between the wall and fluid through a thermal-fluid-structure interaction calculation interface, correcting the temperature field distortion problem caused by neglecting thermal-fluid-structure interaction in traditional simulations. The turbulent viscosity calculation module combines flow field pulsation and thermal convection enhancement terms to achieve dynamic calculation of the turbulent viscosity coefficient. By incorporating state calculations and regionalized model constant adjustments, the limitations of fixed empirical formulas in adapting to the turbulence characteristics of different combustion zones are overcome, significantly improving the accuracy of turbulence simulation. Furthermore, the program can receive combustion parameters in real time, embedding temperature add-ons and real-time turbulent viscosity into the energy evolution module and viscosity term, respectively, achieving deep coupling of multiple physical fields such as combustion, flow, and heat transfer. This allows for the precise capture of characteristics such as pressure pulsations and heat release phase differences in thermoacoustic instability, as well as the generation patterns of pollutants. By establishing a mapping relationship between thermoacoustic instability and pollutant concentration, quantitative data can be provided for combustion system optimization, effectively supporting the refined design of hydrogen-blended natural gas combustion systems. While improving simulation accuracy, this provides a reliable numerical tool for suppressing thermoacoustic instability and reducing pollutant emissions, facilitating the safe and efficient application of hydrogen-blended natural gas in the gas turbine field.

[0125] Corresponding to the aforementioned embodiment of a numerical simulation method for thermal and acoustic instability and pollutant emissions from hydrogen-blended natural gas combustion, this application also provides an embodiment of a numerical simulation apparatus for thermal and acoustic instability and pollutant emissions from hydrogen-blended natural gas combustion.

[0126] Figure 2 This is a schematic diagram of Embodiment 2 of the numerical simulation device for thermoacoustic instability and pollutant emissions during hydrogen-blended natural gas combustion provided in this application. Please refer to... Figure 2 The apparatus provided in this embodiment includes a construction module 210, a calculation module 220, and a simulation module 230;

[0127] The construction module 210 is used to build a gas turbine combustion chamber model;

[0128] The construction module 210 is also used to write a three-dimensional LBM program for the gas turbine combustion chamber model, which is used to simulate the combustion process of multi-component fluids in hydrogen-blended natural gas.

[0129] The calculation module 220 is used to calculate the temperature addition term of the simulated combustion process in real time, and correct the collision equation of the three-dimensional LBM program based on the temperature addition term. The temperature addition term is used to characterize the thermal-fluid-structure coupling state during the combustion process.

[0130] The calculation module 220 is also used to calculate the turbulent viscosity coefficient based on the gas turbine combustion tube model, and the corrected three-dimensional LBM program calculates the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient.

[0131] The simulation module 230 is used by the three-dimensional LBM program to simulate the combustion process inside the gas turbine combustor based on the real-time input combustion parameters and the calculated real-time turbulent viscosity.

[0132] The apparatus of this embodiment can be used to perform... Figure 1 The steps of the method embodiment shown are similar in principle and process, and will not be repeated here.

[0133] The specific implementation process of the functions and roles of each unit in the above device can be found in the implementation process of the corresponding steps in the above method, and will not be repeated here.

[0134] For the device embodiments, since they basically correspond to the method embodiments, the relevant parts can be referred to in the description of the method embodiments. The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this application according to actual needs. Those skilled in the art can understand and implement this without creative effort.

[0135] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A numerical simulation method for the thermoacoustic instability of hydrogen-blended natural gas combustion and pollutant emissions, characterized in that, The method includes: Establish a gas turbine combustion chamber model; A three-dimensional LBM program for the gas turbine combustor model was developed, which was used to simulate the combustion process of multi-component fluids in hydrogen-blended natural gas. The temperature addition term of the combustion process is calculated in real time, and the collision equation of the three-dimensional LBM program is corrected based on the temperature addition term. The temperature addition term is used to characterize the thermal-fluid-structure coupling state during the combustion process. The turbulent viscosity coefficient is calculated based on the gas turbine combustor model, and the corrected three-dimensional LBM program calculates the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient. The three-dimensional LBM program simulates the combustion process inside the gas turbine combustor based on real-time input combustion parameters and calculated real-time turbulent viscosity.

2. The method according to claim 1, characterized in that, The establishment of the gas turbine combustion chamber model includes: Obtain the structural parameters of the combustion chamber of an industrially operating gas turbine; Based on the geometric similarity criterion, the scaling factor is determined using the actual size of the industrial gas turbine combustor as a benchmark. A gas turbine combustor model was established using 3D modeling software based on the structural parameters and the scaling factor.

3. The method according to claim 1, characterized in that, The process of writing the 3D LBM program for the gas turbine combustor model includes: Based on the three-dimensional features and flow field spatial distribution of the gas turbine combustor model, a basic framework for a three-dimensional LBM program is built based on the D3Q19 spatial discretization format. Based on the composition of hydrogen-blended natural gas in the gas turbine combustor model, a multi-component transport and combustion reaction module is added to the basic framework. The multi-component transport and combustion reaction module is used to simulate the coupling process of multi-component mixing and combustion reaction. Based on the wall structure and heat transfer characteristics of the gas turbine combustor model, a thermal-fluid-structure interaction (TFI) calculation interface is generated in the basic framework. The TFI calculation interface is used to simulate the energy transfer at the interface between the wall and the fluid. Based on the turbulent flow characteristics of the gas turbine combustor model, a turbulent viscosity calculation module is generated in the basic framework. The turbulent viscosity calculation module is used to simulate the energy transfer between the energy-efficient vortex and the dissipative vortex. Based on the computational scale and accuracy requirements of the gas turbine combustor model, the parallel computing architecture of the 3D LBM program is adjusted.

4. The method according to claim 1, characterized in that, The temperature additions to the real-time calculated simulation of the combustion process include: Based on the wall thermal boundary conditions of the gas turbine combustor model, the calculation boundary constraints for the temperature addition term are determined. Based on the instantaneous thermophysical parameters of the flow field during the simulated combustion process, the calculation parameters for the additional temperature term are obtained. Based on the boundary constraints and the calculation parameters, a real-time calculation model for the temperature supplement is constructed using the explicit enthalpy method, and the temperature supplement is calculated based on the real-time calculation model.

5. The method according to claim 1, characterized in that, The calculation of the turbulent viscosity coefficient based on the gas turbine combustor model includes: Model parameters of the gas turbine combustor model are extracted, and flow field data is collected; wherein, the flow field data includes particle distribution function, equilibrium particle distribution function, temperature gradient, and energy transfer characterization vector at each grid point; Calculate the flow field fluctuation value based on the particle distribution function and the equilibrium particle distribution function; The turbulence enhancement term induced by thermal convection is calculated based on the temperature gradient and the energy transfer characterization vector. The turbulent viscosity coefficient is calculated based on the flow field fluctuation value and the turbulence enhancement term.

6. The method according to claim 1, characterized in that, The modified three-dimensional LBM program calculates the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient; including: The variables for calculating real-time turbulent viscosity are determined based on the turbulent viscosity calculation formula in the three-dimensional LBM program, and the variables include at least model constants; The model constants are adjusted according to the flow field characteristics of different combustion zones within the gas turbine combustor. In each time step iteration of the 3D LBM program, the real-time turbulent viscosity of each grid point is directly calculated by combining the adjusted model constants.

7. The method according to claim 1, characterized in that, The three-dimensional LBM program simulates the combustion process inside the gas turbine combustor based on real-time input combustion parameters and calculated real-time turbulent viscosity; including: Read the real-time input combustion parameters, convert the combustion parameters into the initial conditions of the particle distribution function at the inlet boundary, load the mesh parameters and wall thermal boundary conditions of the gas turbine combustor model, and complete the simulation initialization; The temperature addition term is embedded as a source term into the energy evolution module of the LBM collision equation to correct the spatial distribution of the temperature field. The real-time turbulent viscosity is embedded into the viscous term of the LBM collision equation to correct the momentum exchange during particle collisions. The combustion reaction rate is calculated based on real-time combustion parameters; By combining the corrected viscosity term and the temperature addition term, the particle collision and migration process is executed to complete the simulation of the combustion process.

8. The method according to claim 1, characterized in that, The method for writing the 3D LBM program for the gas turbine combustor model also includes: Based on the chemical reaction mechanism of hydrogen-doped natural gas, the component concentrations of multiple target components in the reaction source term are determined, and the reaction rates of each elementary reaction in the hydrogen-doped natural gas are calculated. The net generation rate of each target component is calculated based on the reaction rate and the component concentration. The net generation rate is transformed into a reaction source term form suitable for the evolution of the LBM probability density function. The reaction source term is then embedded into the collision equation of the probability density function of each component to simulate the coupling between the combustion reaction and the flow field.

9. The method according to claim 1, characterized in that, After the three-dimensional LBM program simulates the combustion process inside the gas turbine combustor based on real-time input combustion parameters and calculated real-time turbulent viscosity, it includes: Thermoacoustic instability parameters and pollutant concentrations were extracted from the simulation results. A pollutant mapping relationship is established based on the aforementioned thermoacoustic instability characteristic parameters and pollutant concentration, and the combustion system is optimized based on the pollutant mapping relationship.

10. A numerical simulation device for thermoacoustic instability and pollutant emissions during the combustion of hydrogen-blended natural gas, characterized in that, The device includes a construction module, a computing module, and a simulation module; wherein... The construction module is used to build a gas turbine combustion chamber model; The construction module is also used to write a three-dimensional LBM program for the gas turbine combustor model, which is used to simulate the combustion process of multi-component fluids in hydrogen-blended natural gas. The calculation module is used to calculate the temperature additions of the simulated combustion process in real time, and to correct the collision program of the three-dimensional LBM program based on the temperature additions. The temperature additions are used to characterize the thermal-fluid-structure coupling state during the combustion process. The calculation module is also used to calculate the turbulent viscosity coefficient based on the gas turbine combustor model, and the corrected three-dimensional LBM program calculates the real-time turbulent viscosity of the combustion process based on the turbulent viscosity coefficient. The simulation module is used by the three-dimensional LBM program to simulate the combustion process inside the gas turbine combustor based on the real-time input combustion parameters and the calculated real-time turbulent viscosity.

Images