A high-precision simulation method for an aeroengine combustion chamber

Abstract

The present invention provides a high-precision simulation method for an aeroengine combustion chamber, comprising the following steps: Step 1: Input the fuel and air flow rates at the inlet of the aeroengine combustion chamber; Step 2: Simulate the atomization process, evaporation process and combustion process in the combustion chamber; Step 3: Determine whether the residual iteration at the combustion chamber outlet is completed. If so, proceed to the next step; otherwise, return to Step 2; Step 4: Output the parameters at the outlet of the aeroengine combustion chamber. The beneficial effects of the present invention are as follows: The present invention can achieve a substantial improvement in the overall accuracy of atomization combustion, has high-precision characteristics, and does not require parameter adjustment steps according to specific atomization combustion conditions. It has strong versatility and can meet the high-precision requirements in the iterative optimization of aeroengine combustion chambers.

Description

Technical Field

[0001] The present invention relates to the technical field of aeroengines or gas turbines, and particularly to a high-precision simulation method for an aeroengine combustion chamber. Background Art

[0002] The atomization combustion process inside an engine combustion chamber includes complex processes such as atomization of liquid fuel, evaporation, mixing of vapor and air, and combustion. Nowadays, the overall simulation of atomization combustion simplifies the atomization process. For example, the Lagrangian method is used to describe the atomization process. The liquid fuel is not injected in the form of a liquid column but is simplified into a large number of fully atomized droplets [1]. Each droplet is characterized by a Lagrangian point. We call this type of model the first type of atomization model. Furthermore, researchers have developed typical atomization models to describe the atomization process, such as the Taylor Analogy Breakup (TAB) breakup model proposed by O’Rourke and Amsden in the 1980s [2], the WAVE model proposed by Reitz [3], the Kelvin-Helmholtz Rayleigh-Taylor (KH-RT) model proposed by Beale et al. [4], and the stochastic breakup model proposed by Apte [5], etc. The TAB model is based on the Taylor similarity theory and analogizes the oscillation and deformation of droplets with an elastic particle system. The WAVE model assumes that the KH instability is the dominant mechanism for atomization, and the KH-RT model additionally considers the RT instability caused by interface acceleration. The stochastic breakup model assumes that the size and number density of droplets follow a probability density function distribution. We call these models, which are controlled by a large number of artificial parameters and assume the atomization mechanism in advance, the second type of atomization model.

[0003] In fact, there is no unified conclusion on the evolution mechanism of atomization. Using the above two types of atomization models with strong assumptions will seriously affect the overall simulation accuracy of atomization combustion. For example, Som and Aggarwal [6] studied the influence of different atomization models (KH and KH-ACT models) on combustion characteristics and found that the prediction differences of different atomization models for the atomization penetration distance reached 33%, and the prediction difference for the flame lift-off height even reached 50%. Greenberg [7] found that using different assumptions of droplet size distribution would cause flame oscillation. Esclapez et al. [8] performed large-eddy simulation on the atomization combustion process of a real aeroengine combustion chamber and found that the downstream droplet velocity obtained by using the stochastic atomization model differed from the experimental value by nearly 100%. The two types of atomization models developed in the 1980s and 1990s have been used until now, and their limitations have severely restricted the accuracy of atomization combustion simulation. Improving the accuracy of atomization simulation has become an urgent problem to be solved for the overall simulation accuracy of atomization combustion.

[0004] At present, the simulation accuracy of an aero-engine combustion chamber is limited, and the overall simulation of atomization combustion inside the combustion chamber simplifies the atomization process. Although the current simulation of atomization combustion can generally meet some of the requirements in engineering, a large number of studies have shown that for unsteady phenomena such as liquid spray auto-ignition and point extinction, due to the low simulation accuracy of the atomization process, the overall prediction accuracy of atomization combustion is limited and cannot meet the engineering requirements. Summary of the Invention

[0005] The present invention provides a high-precision simulation method for an aero-engine combustion chamber, including the following steps:

[0006] Step 1: Input the fuel and air flow rates at the inlet of the aero-engine combustion chamber.

[0007] Step 2: Simulate the atomization process, evaporation process, and combustion process inside the combustion chamber.

[0008] Step 3: Determine whether the residual iteration at the combustion chamber outlet is completed. If so, proceed to the next step; otherwise, return to Step 2.

[0009] Step 4: Output the parameters at the outlet of the aero-engine combustion chamber.

[0010] As a further improvement of the present invention, in Step 2, the LevelSet method is used to track the gas-liquid interface during the simulation of the atomization process inside the combustion chamber.

[0011] As a further improvement of the present invention, in the Level Set method, the gas-liquid interface is represented as the zero isosurface of a smooth function φ(x,t), and this smooth function is usually taken as the signed distance function:

[0012] |G(x,t)| = |x - x Γ |,

[0013] where t is the time, and x Γ is the point on the interface closest to x;

[0014] The motion of the phase interface is determined by the following transport equation:

[0015]

[0016] where u is the fluid velocity; sub-grid interface characterization is adopted, that is, the grid is refined only at the phase interface;

[0017] The semi-Lagrangian transport format is adopted, and the semi-Lagrangian format is based on the following fact:

[0018] The scalar G should remain unchanged along the trajectory of the material point under the action of the transport velocity u. At time t n+1 through the position xn+1 The trajectory can be traced backward to time t n = t n+1 -Δt, so as to obtain the position point x at the previous moment n , thus at position x n+1 the LS function value G n+1 is obtained in the following form:

[0019] G n+1 (x n+1 ) = G n (x n )

[0020] Adopt the Lagrangian transport format and solve an ordinary differential equation between x n and x n+1 .

[0021] As a further improvement of the present invention, the smooth function takes a positive value in the liquid phase, a negative value in the gas phase, and a value of 0 at the interface.

[0022] As a further improvement of the present invention, the LevelSet method needs to be re-initialized.

[0023] As a further improvement of the present invention, only 8 nearest grid points are needed for the re-initialization, and the velocity u and the LS (level set) scalar G are interpolated between the orthogonality points by trilinear interpolation.

[0024] As a further improvement of the present invention, the re-initialization is performed every 100 time steps.

[0025] As a further improvement of the present invention, in step 2, the simulation of the evaporation process in the combustion chamber includes:[[]]

[0026] During the atomization process of aviation liquid fuel in the combustion chamber and subsequent stages, along with the evaporation process of the liquid fuel, solve the Navier-Stokes equations of variable density and low Mach number:

[0027]

[0028] where ρ, p and μ represent density, velocity, pressure and dynamic viscosity respectively;

[0029] The evaporation of the liquid droplets is controlled by the temperature and component concentration at the interface, so the conservation equations are used to solve the temperature and mass fraction of each component of the whole field, as follows:

[0030]

[0031] where T, Y, C p , λ and Dm respectively represent temperature, mass fraction, specific heat capacity at constant pressure, thermal conductivity, and mass diffusion coefficient;

[0032] For the step condition at the interface, it refers to the Rankine - Hugoniot step, which can be considered as the conservation of energy and mass on both sides of the interface, as shown in the following equation:

[0033]

[0034] where h lg represents the latent heat of vaporization. Considering that the gradient of the mass fraction in the liquid phase is zero, the solution of the evaporation rate can be simplified as:

[0035]

[0036] where represents the vapor concentration at the gas - side interface, which can be obtained through the Clausius - Clapeyron relation:

[0037]

[0038]

[0039] where is the vapor pressure at the interface, p atm is the ambient pressure, m vap is the molar mass of the vapor, R is the ideal gas constant, T Γ is the interface temperature, T B is the boiling temperature of the liquid under the ambient pressure, m g is the molar mass of the gas phase;

[0040] The virtual fluid method is used to handle the step condition. The entire computational domain uses a uniform staggered grid, that is, scalars are defined at the grid centers while velocity components are stored on the grid faces. Considering the discontinuity of the velocity field at the interface, it is necessary to determine the virtual velocity values on the virtual grids, that is, the extension of the gas / liquid phase velocities in the liquid / gas phase computational domains, which can be specifically expressed as:

[0041]

[0042] As a further improvement of the present invention, in the step 2, the simulation of the combustion process in the combustion chamber includes:

[0043] The gas - phase combustion is described by the Navier - Stokes equations with variable density and low Mach number, where the influence of the liquid spray is a mass source term in the continuity equation and a momentum source term in the momentum equation The equations are expressed as

[0044]

[0045] In addition, according to the single-step chemical reaction mechanism, the scalar transport equations for the mass fractions of fuel (F), oxidizer (O), and product (P) need to be solved, including the chemical reaction source term and the source term of the liquid fuel in the continuity equation

[0046]

[0047] Finally, the transport equation for the gas-phase temperature needs to be solved:

[0048]

[0049] where, is the energy exchange rate with the droplet, is the heat release rate of combustion, the heat capacity c P and the viscosity μ are kept constant, and the Lewis number for each component is 1.

[0050] As a further improvement of the present invention, in the said step 4, the temperature and pollutant parameters at the outlet of the aero-engine combustion chamber are output.

[0051] The beneficial effects of the present invention are as follows: The present invention can achieve a substantial improvement in the overall accuracy of atomization combustion, has high-precision characteristics, and does not require steps for parameter adjustment according to specific atomization combustion conditions, has strong versatility, and can meet the high-precision requirements in the iterative optimization of aero-engine combustion chambers. Description of the Drawings

[0052] Figure 1 is the flow chart of the high-precision simulation method of the present invention. Detailed Embodiments

[0053] The present invention discloses a high-precision simulation method for an aero-engine combustion chamber, which uses a high-precision interface tracking algorithm to accurately simulate the atomization process and couples the evaporation and combustion processes to achieve an overall high-precision simulation of the atomization combustion process.

[0054] The high-precision simulation method for an aero-engine combustion chamber disclosed by the present invention includes the following steps:

[0055] Step 1: Input the fuel and air flow rates at the inlet of the aero-engine combustion chamber.

[0056] Step 2: Simulate the atomization process, evaporation process, and combustion process in the combustion chamber.

[0057] Step 3: Determine whether the residual iteration at the combustion chamber outlet is completed. If so, proceed to the next step; otherwise, return to Step 2.

[0058] Step 4: Output the parameters at the outlet of the aero-engine combustion chamber.

[0059] In the said Step 2, it specifically further includes:

[0060] 1. Simulation of the atomization process in the combustion chamber. The atomization process in the combustion chamber is the result of the comprehensive action of aviation liquid fuel and the air in the combustion chamber. The key to the high-precision simulation of this process lies in the tracking of the gas-liquid interface. The present invention uses the LevelSet method to track the gas-liquid interface. In the LevelSet method, the gas-liquid interface is represented as the zero isosurface of a smooth function φ(x, t), and this smooth function is usually taken as the signed distance function:

[0061] |G(x, t)| = |x - x Γ |,

[0062] where t is the time, and x Γ is the point on the interface closest to x. This function takes positive values in the liquid phase, negative values in the gas phase, and a value of 0 on the interface. The movement of the phase interface is determined by the following transport equation,

[0063]

[0064] where u is the fluid velocity. The most typical LS transport format is the WENO format, which can ensure the accuracy and numerical stability of the transport. However, WENO is prone to excessive dissipation in the transport of small-scale scalar structures, which will cause serious problems for the gas-liquid two-phase flow to be studied. Since a large number of small-scale liquid structures are generated during the actual atomization process, this poses a huge challenge to the grid quantity. The present invention uses sub-grid interface characterization, that is, only the grid is refined at the phase interface, thereby greatly reducing the computational amount. This method is also called the pseudo-spectral sub-grid reconstruction method.

[0065] The semi-Lagrangian transport format is an efficient and accurate scalar transport format that can avoid the serious CFL limit caused by the rapidly decreasing distance between positive intersection points. Without discretizing the transport equation, the semi-Lagrangian format is based on the following fact: the scalar G should remain unchanged along the trajectory of the material point under the action of the transport velocity u. At time t n+1 through the position x n+1 the trajectory can be traced backward to time t n = t n+1 -Δt, so as to obtain the position point x n at the previous moment. Therefore, the LS function value G n+1 at the position x n+1 can be obtained in the following form: G n+1 (x n+1) = G n (x n ). Since it is a Lagrangian transport format, a relatively large time step can be adopted. At the same time, it is necessary to solve an ordinary differential equation (ODE) between x n and x n+1 . The phase interface transport equation is discretized by the fourth-order Runge-Kutta method.

[0066] For the LevelSet method, a re-initialization process is required. The re-initialization only needs to use 8 nearest grid points, and the velocity u and the LS (level set) scalar G are interpolated between the orthogonality points by trilinear interpolation. After verification, it is not necessary to perform the re-initialization process at each step. We perform the re-initialization every 100 time steps.

[0067] 2. Simulation of the evaporation process in the combustion chamber. During the atomization process of aviation liquid fuel in the combustion chamber and in the subsequent stage, along with the evaporation process of the liquid fuel, the present invention solves the Navier-Stokes equations of variable density and low Mach number:

[0068]

[0069] where ρ, p, and μ represent density, velocity, pressure, and dynamic viscosity respectively.

[0070] The evaporation of the droplet is controlled by the temperature and component concentration at the interface. Therefore, the conservation equations are used to solve the temperature and mass fractions of each component of the whole field, as follows:

[0071]

[0072] where T, Y, C p , λ, and D m represent temperature, mass fraction, specific heat capacity at constant pressure, thermal conductivity, and mass diffusion coefficient respectively.

[0073] In addition, the transport equation for capturing the interface using the Level Set method used in the present invention needs to consider the evaporation effect:

[0074]

[0075] where and ρ l represent the hyperbolic tangent function, evaporation rate, and liquid phase density respectively.

[0076] The important step condition at the interface refers to the Rankine-Hugoniot step, which can be considered as the conservation of energy and mass on both sides of the interface, as shown in the following equation,

[0077]

[0078] where h lg represents the latent heat of vaporization. Considering that the gradient of the mass fraction in the liquid phase is zero, the solution of the evaporation rate can be simplified as:

[0079]

[0080] where represents the vapor concentration at the gas-phase interface. can be obtained through the Clausius-Clapeyron relation:

[0081]

[0082] where is the vapor pressure at the interface, p atm is the ambient pressure, m vap is the molar mass of the vapor, R is the ideal gas constant, T Γ is the interface temperature, T B is the boiling temperature of the liquid under the ambient pressure, m g is the molar mass of the gas phase. It should be noted that the gas-phase and liquid-phase vapors are different.

[0083] The accurate discretization of the boundary conditions at the interface is of great significance in this method. The present invention uses the virtual fluid method to handle these step conditions. The entire computational domain uses a uniform staggered grid, that is, the scalar is defined at the grid center and the velocity components are stored on the grid faces. Considering the discontinuity of the velocity field at the interface, we need to determine the virtual velocity values on the virtual grid, that is, the extension of the gas / liquid-phase velocity in the liquid / gas-phase computational domain. Specifically, it can be expressed as:

[0084]

[0085] 3. Simulation of the combustion process in the combustion chamber. The gas-phase combustion is described by the Navier-Stokes equations with variable density and low Mach number, where the influence of the liquid spray is the mass source term in the continuity equation and the momentum source term in the momentum equation The equations are expressed as

[0086]

[0087] In addition, according to the one-step chemical reaction mechanism, the scalar transport equations for the mass fractions of the fuel (F), oxidizer (O), and product (P) need to be solved, including the chemical reaction source term and the source term of the liquid fuel in the continuity equation

[0088]

[0089] Finally, the transport equation for the gas phase temperature needs to be solved:

[0090]

[0091] wherein, is the energy exchange rate with the droplet, and is the heat release rate of combustion. The heat capacity c P and the viscosity μ are kept constant, and the Lewis number for each component is 1.

[0092] The beneficial effects of the present invention are as follows: The present invention can achieve a substantial improvement in the overall accuracy of atomization combustion, has high-precision characteristics, and does not require the step of parameter adjustment according to specific atomization combustion conditions, has strong versatility, and can meet the high-precision requirements in the iterative optimization of aeroengine combustors.

[0093] The above content is a further detailed description of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is only limited to these descriptions. For those of ordinary skill in the technical field to which the present invention pertains, without departing from the concept of the present invention, several simple deductions or substitutions can still be made, and all should be regarded as belonging to the protection scope of the present invention.

Claims

1. A high-precision simulation method for an aero-engine combustion chamber, characterized in that It includes the following steps: Step 1: Input the fuel and air flow rates at the inlet of the aero-engine combustion chamber; Step 2: Simulate the atomization process, evaporation process, and combustion process in the combustion chamber; Step 3: Judge whether the residual iteration at the combustion chamber outlet is completed. If so, proceed to the next step; otherwise, return to Step 2; Step 4: Output the parameters at the outlet of the aero-engine combustion chamber; In Step 2, the Level Set method is used to track the gas-liquid interface in the simulation of the atomization process in the combustion chamber; In the Level Set method, the gas-liquid interface is represented as the zero isosurface of a smooth function φ(x,t), and this smooth function is usually taken as the signed distance function: |G(x,t)| = |x - x Γ |, where t is time and x Γ is the point on the interface that is the shortest distance from x; The motion of the phase interface is determined by the following transport equation: where u is the fluid velocity; sub-grid interface characterization is adopted, that is, the grid is refined only at the phase interface; The semi-Lagrangian transport format is adopted, and the semi-Lagrangian format is based on the following fact: The scalar G should remain constant along the trajectory of the material point under the action of the transport velocity u. n+1 By position x n+1 The trajectory can be traced back to time t n =t n+1 -Δt, thus obtaining the position point x at the previous moment n , so at position x n+1 The LS function value G n+1 Obtained through the following forms: G n+1 (x n+1 ) = G n (x n ) Adopt a Lagrangian transport format and solve an ordinary differential equation between x n and x n+1 .

2. The high-precision analog method according to claim 1, wherein The smooth function takes positive values in the liquid phase, negative values in the gas phase, and 0 on the interface.

3. The high-precision analog method according to claim 1, characterized in that Re-initialization is required for the Level Set method.

4. The high-precision simulation method according to claim 3, wherein For the re-initialization, only 8 nearest grid points are needed, and the velocity u and the LS scalar G are interpolated between the orthogonality points using trilinear interpolation.

5. The high-precision analog method according to claim 3, characterized in that, Re-initialization is performed every 100 time steps.

6. The high-precision analog method according to claim 1, characterized in that In Step 2, the simulation of the evaporation process in the combustion chamber includes: During the atomization process of the aviation liquid fuel in the combustion chamber and subsequent stages, along with the evaporation process of the liquid fuel, solve the Navier-Stokes equations with variable density and low Mach number: where ρ, p, and μ represent density, velocity, pressure, and dynamic viscosity, respectively; The evaporation of the liquid droplets is controlled by the temperature and component concentration at the interface. Therefore, the conservation equations are used to solve the temperature and mass fractions of each component for the whole field, as follows: where T, Y, C p , λ, and D m represent temperature, mass fraction, specific heat capacity at constant pressure, thermal conductivity, and mass diffusivity, respectively; The step condition at the interface refers to the Rankine-Hugoniot step, which can be considered as the conservation of energy and mass on both sides of the interface, as shown in the following equation, where h lg represents the latent heat of evaporation. Considering that the gradient of the mass fraction in the liquid phase is zero, the solution of the evaporation rate can be simplified as follows: where represents the vapor concentration at the gas-phase side interface, which can be obtained by the Clausius-Clapeyron relation: where is the vapor pressure at the interface, p atm is the ambient pressure, m vap is the molar mass of the vapor, R is the ideal gas constant, T Γ is the interface temperature, T B is the boiling temperature of the liquid at the ambient pressure, m g is the molar mass of the gas phase; The virtual fluid method is adopted to handle the step condition. The entire computational domain uses a uniform staggered grid, that is, the scalar is defined at the grid center and the velocity components are stored on the grid faces. Considering the discontinuity of the velocity field at the interface, it is necessary to determine the virtual velocity values on the virtual grid, that is, the extension of the gas / liquid phase velocity in the liquid / gas phase computational domain, which can be specifically expressed as:

7. The high-precision simulation method according to claim 1, wherein In Step 2, the simulation of the combustion process in the combustion chamber includes: Gas-phase combustion is described by the Navier-Stokes equations with variable density and low Mach number, where the influence of the liquid spray is a mass source term in the continuity equation and a momentum source term in the momentum equation The equations are expressed as In addition, according to the single-step chemical reaction mechanism, the scalar transport equations for the mass fractions of fuel (F), oxidizer (O), and product (P) need to be solved, including the chemical reaction source term and the source term of the liquid fuel in the continuity equation Finally, it is necessary to solve the transport equation of the gas-phase temperature: where, for the energy exchange rate with the droplet, and for the heat release rate of combustion, the heat capacity c P and the viscosity μ are kept constant, and the Lewis number for each component is 1.

8. The high-precision simulation method according to claim 1, wherein In Step 4, output the temperature and pollutant parameters at the outlet of the aero-engine combustion chamber.

Images