A numerical calculation method for high-speed jet flow in the nozzle swing of underwater solid rocket motor
Through the RANS equation and Realizable k-epsilon turbulence mode combined with near-wall area processing, partition grid and finite volume method are used to solve the problem of boundary layer separation and shock interference in the flow field calculation of supersonic jet elements, and the complexity is achieved High-precision simulation and conservation calculation of flow field.
Patent Information
- Application Number
- CN202210512362.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-05-12
- Publication Date
- 2024-11-08
- Estimated Expiration
- 2042-05-12
AI Technical Summary
In the prior art, the flow field calculation of the supersonic jet element is difficult to accurately reflect the boundary layer separation phenomenon and the mutual interference between the shock wave and the boundary layer, and the single-domain grid generation method cannot effectively solve the problem of complex configuration flow fields.
Using RANS equation combined with Realizable k-epsilon turbulence mode, near-wall area processing, enclosing equation and initial conditions, numerical calculations are carried out through partition grid generation and finite volume method to realize the high-speed jet flow of the nozzle of the underwater solid rocket engine Accurate simulation.
It improves the accuracy and conservation of flow field calculations, is suitable for calculations of complex configuration flow fields and transsonic flow, reduces computer storage requirements, is suitable for parallel computing, and can easily describe distorted areas.
Smart Images

Figure CN114996866B_ABST
Abstract
Description
Technical Field
[0001] The invention discloses a numerical calculation method for high-speed jet flow of an underwater solid rocket engine nozzle swing, and belongs to the technical field of jet pulse engines. Background Art
[0002] The structure of supersonic jet elements is complex, and it is difficult to generate high-quality single-domain structure calculation grids. The internal flow field is a complex viscous flow field. The use of unstructured grids has a great influence on the calculation accuracy of the boundary layer. Unstructured grids are used in the prior art, but the calculation results are not ideal and cannot well reflect the boundary layer separation phenomenon near the control port and the mutual interference phenomenon between the shock wave and the boundary layer. Now, a numerical calculation method for the swinging high-speed jet flow of an underwater solid rocket engine nozzle is urgently needed to solve the above-mentioned problems. Summary of the invention
[0003] In view of the shortcomings of the prior art, the purpose of the present invention is to provide a numerical calculation method for the high-speed jet flow of the swinging nozzle of an underwater solid rocket engine, so as to solve the problems raised in the above-mentioned background technology. The present invention has a reasonable structure and solves the flow field calculation problems of complex configurations that cannot be solved or have poor effects using single-domain grid generation methods.
[0004] In order to achieve the above object, the present invention is implemented through the following technical scheme: a numerical calculation method for the high-speed jet flow of the nozzle swing of an underwater solid rocket engine, comprising the following steps:
[0005] Step 1: Establish a mathematical model, using the RANS equation as the control equation, combined with the Realizable k-epsilon turbulence model, near-wall area treatment, closure equation, boundary conditions and initial conditions as the mathematical model for the numerical calculation of supersonic jet elements;
[0006] Step 2: Mesh the mathematical model and fluid calculation domain established in step 1, establish a fluid calculation domain for the underwater solid rocket engine nozzle geometry model obtained in step 1, and perform geometric optimization on the interface between the two fluid calculation domains. Transfer data between the calculation domains to simulate the nozzle swing condition, mesh the flow domain in blocks, and perform mesh independence verification to obtain the optimal number of grids.
[0007] Step 3: Discretization and solution of control equations;
[0008] Step 4: Based on steps 1 to 3, numerical calculation and analysis of the flow characteristics of the underwater solid rocket engine nozzle swing high-speed jet are carried out;
[0009] In step 1, a mathematical model including control equations, turbulence patterns, turbulence patterns in the near-wall region, and improved wall functions is established;
[0010] S1. Control equations Based on the flow characteristics inside the supersonic jet element, the volume force and heat source are ignored, and the wall is assumed to be an adiabatic wall. In the following equations, u i Replace its time mean u i ' is the pulsation value, in the Cartesian coordinate system:
[0011] (1) Continuity equation
[0012]
[0013] (2) Momentum equation
[0014]
[0015] In the formula, Reynolds stress Its expression will be given in the turbulence model, τ ij is the viscous stress tensor, which can be expressed as:
[0016] Strain rate of fluid motion—
[0017] Volume expansion rate of a particle— δ ij is the Kronecker symbol; μ' is the "volumetric viscosity coefficient" or "second viscosity coefficient", which reflects the viscous stress of the fluid deviating from the thermodynamic pressure caused by the volume change. Except for extreme cases such as high temperature and high-frequency sound waves, μ' can be approximately considered to be 0 in general gas motion;
[0018] (3) Energy equation
[0019]
[0020] Where:
[0021] J j' is the diffusion amount of the mixed gas component j'
[0022] in, c p,j' is the constant pressure specific heat of the mixed gas component j', take T ref =298.15K;
[0023] λ is the thermal conductivity, c p is the specific heat at constant pressure, Pr t =0.85
[0024]
[0025] Effective turbulent eddy viscosity coefficient μeff and the turbulent eddy viscosity coefficient μ t Its expression will be given in the turbulence model;
[0026] S2. In the turbulence mode, there is a strong adverse pressure gradient and flow separation in the flow boundary layer of the supersonic jet element, and under certain conditions, there is a large backflow. The Realizable k-epsilon mode is used to calculate the turbulence. The definitions of turbulent kinetic energy k and turbulent dissipation rate ε are as follows:
[0027]
[0028] (1) Transport equation of turbulent kinetic energy k
[0029]
[0030] Where:
[0031] G k =μ t S 2 , the average strain rate tensor coefficient
[0032] Y M =2ρεM t 2 , turbulent Mach number a is the local speed of sound
[0033] (2) Transport equation of turbulent dissipation rate ε
[0034]
[0035] Where:
[0036] And η=Sk / ε
[0037] (3) Turbulent eddy viscosity coefficient μ t
[0038] At high Reynolds numbers, the turbulent eddy viscosity coefficient μ t Expressed as
[0039]
[0040] Where:
[0041] and It is related to the rotating reference surface and is not considered in this calculation;
[0042] A0=4.04,
[0043]
[0044] At low Reynolds numbers, the effective turbulent eddy viscosity coefficient μ eff It can be obtained by the following formula
[0045]
[0046] Where: C υ ≈100
[0047] (4) Reynolds stress
[0048] According to the Boussinesq hypothesis, the Reynolds stress The relationship between the time-averaged velocity gradient is
[0049]
[0050] (5) Model constant
[0051] C2=1.9,σ k =1.0,σ ε =1.2;
[0052] S3. In the turbulence model near the wall, a wall treatment method combining a two-layer wall model with an improved wall function is used: in the two-layer near-wall model, the entire flow area is divided into a viscous influence area and a complete turbulence area according to the size of the turbulence Reynolds number; the turbulence Reynolds number Re y is defined as follows:
[0053] y is the normal distance from the center of the computational grid cell to the wall;
[0054] In the fully turbulent region Use Realizable k-epsilon turbulence model;
[0055] In the near-wall region affected by viscosity The Wolfstein equation model is used. In this model, the transport equation of turbulent kinetic energy k is the same as that of the Realizable k-epsilon turbulence model, and the turbulent eddy viscosity coefficient μ t,2layer and the turbulent dissipation rate ε t,2layer The calculation of is as follows:
[0056] ε t,2layer =k 1.5 / l ε
[0057] In the above formula:
[0058] μ tis the turbulent eddy viscosity coefficient at high Reynolds numbers, and its calculation is shown in the following formula:
[0059] l μ and l ε is the length scale and is calculated as follows:
[0060] l μ =c l y[1-exp(-Re y / A μ )]
[0061] l ε =c l y[1-exp(-Re y / A ε )]
[0062] The constant in the above formula is A μ =70, A ε =2c l
[0063] In order to make the two-layer model defined μ t,2layer It can be defined with the outer layer high Reynolds number μ t Smooth transition, defining an improved turbulent eddy viscosity coefficient μ t,enh , which is defined as follows
[0064] μ t,enh =λ ε μ t +(1-λ ε )μ t,2layer
[0065] Where λ ε is a compromise function, which is used to prevent the convergence of the solution from being blocked when the k-ε solution of the outer layer does not match the solution calculated by the two-layer model. It is defined as follows:
[0066]
[0067] In the above formula, the constant A determines the trade-off function λ ε width; if the variable △Re is given y , to make λ ε The value of A should not exceed 1% of its far-field value. And △Re y The value can be found in Given between 5% and 20% of the value;
[0068] S4. The improved wall function is a hybrid function that organically combines the linear wall law and the logarithmic wall law. For the average velocity, the improved wall function can be expressed as:
[0069]
[0070] In the above formula, Where a = 0.01c, b = 5 / c, and
[0071] Taking the derivative of the formula, we get
[0072]
[0073] The improved turbulent wall law can be expressed as
[0074]
[0075] in
[0076]
[0077] This is the position where the slope of the logarithmic law remains constant, and its value is 60 in the calculation;
[0078]
[0079] From the above expressions of α, β and γ, it can be seen that the improved turbulent wall law takes into account the effects of pressure gradient and heat;
[0080] The improved laminar wall law can be expressed as
[0081]
[0082] The improved temperature wall function is expressed as
[0083]
[0084] in, Pr is the molecular Prandtl number;
[0085] S5. Closed-form equations
[0086] In order to make the above equation closed, the following closed equation must be added:
[0087] State equation: f(p,ρ,T)=0
[0088] The density of the flow field is calculated using the complete gas state equation, that is: ρ = p / RT;
[0089] Specific heat at constant pressure:
[0090] R is the universal gas constant, M is the molecular weight of the gas, and f is the number of degrees of freedom of molecular motion. For polyatomic molecules, f=7;
[0091] Thermal conductivity:
[0092] The molecular viscosity coefficient μ is calculated using the Sutherland formula when the temperature is neither too high nor too low, that is,
[0093]
[0094] μ0 is the viscosity coefficient at the reference temperature T0, S is the Sutherland constant;
[0095] For air (210K~1900K), μ0=1.716×10 -5 Pa﹒ s, T0=273K, S=111K;
[0096] When the temperature is low, the Keyes formula is used for calculation, that is,
[0097]
[0098] S6. The supersonic jet element consists of three inlets, two outlets and a wall. Among the three inlets, one is the main gas source inlet and the other two are control flow inlets. According to the flow characteristics of the supersonic jet element and the definition method of boundary conditions in the FLUENT software, the main gas source inlet boundary condition is defined as the pressure inlet boundary condition, the two control flow inlet boundary conditions are defined as the mass inflow boundary conditions, and the two outlet boundary conditions are defined as the pressure outlet boundary conditions. For steady or unsteady flow, when the time-marching algorithm is used to solve, in addition to the boundary conditions, the initial conditions also need to be given. The following is the formulation of each boundary condition and initial condition when calculating the steady or unsteady flow inside the supersonic jet element and the corresponding mathematical treatment;
[0099] (1) The total pressure (P0), total temperature (T0), flow direction, turbulence intensity and hydraulic diameter are given at the pressure inlet boundary conditions. In addition, the static pressure (Ps) is also given to initialize the flow field. From the isentropic relationship of ideal gas, we can get
[0100]
[0101] In the above formula
[0102]
[0103] Where c is the speed of sound, k = cp / cv is the isentropic index,
[0104] The density can be obtained from the ideal gas state equation, that is,
[0105]
[0106] The static temperature Ts in the above formula can be calculated by the following formula
[0107]
[0108] The density, static temperature and velocity scalar values at the pressure inlet can be obtained, and then the velocity component values at the pressure inlet in each direction can be obtained according to the given flow direction;
[0109] (2) Mass inflow boundary conditions
[0110] Gives the mass flow rate Total temperature (T0), static pressure (Ps), flow direction, turbulence intensity and hydraulic directivity
[0111] The mass flow rate is calculated as follows:
[0112]
[0113] In the above formula, vn is the normal velocity and A is the inlet area.
[0114] The density at the inlet is obtained from the ideal gas state equation, that is,
[0115]
[0116] In the above formula, if the flow at the inlet is supersonic, Ps uses a given static pressure value; if the flow at the inlet is subsonic, Ps is obtained by extrapolating the grid center value of the computational grid at the inlet;
[0117] The static temperature in the formula can be calculated by the following formula
[0118]
[0119] In the mass inflow boundary condition, the flux of the corresponding variable at the inlet can be obtained by the normal velocity vn and the inlet value of the variable. The mass flux is ρv n , the flux of turbulent kinetic energy is ρkv n .
[0120] Furthermore, the grid is divided into supersonic jet element grid partitions and boundary layer grids. The entire calculation area is divided into ten subdomains, and the grid of each subdomain is generated separately. Among them, subdomain one adopts an unstructured grid, subdomain eight adopts a hybrid grid in which a structured grid and an unstructured grid coexist, and the remaining eight subdomains adopt a structured grid. When generating calculation grids for each subdomain, the grids at the connection of each subdomain are made completely continuous as much as possible, and no interpolation calculation is required, so as to reduce the amount of information exchange at the regional boundary and make the boundary processing simple. In this way, in the process of capturing cross-region shock waves in the flow field inside the jet element, because the variable transfer at the regional boundary is conserved, the shock wave can be guaranteed to be continuous across regions. According to the flow characteristics near the walls in the supersonic jet element, by giving appropriate h1, m and R values, a boundary layer grid with a density roughly consistent with the change gradient of the flow field parameters can be generated.
[0121] Furthermore, in step three, the control equations are discretized and solved. The flow inside the supersonic jet element is a transonic viscous internal flow, and there is a twisted area. Therefore, the finite volume method is used to discretize the control equations into algebraic equations that can be numerically calculated on the physical space curve grid;
[0122] The finite volume method is described by the control equation in integral form. Therefore, for any two-dimensional control volume S, the control equation is converted into a vector equation in integral form:
[0123]
[0124] In the formula, and They are defined as:
[0125]
[0126] Where l is the loop length of the control volume S, is the velocity vector, u and v are the velocity components in Cartesian coordinates, E is the total energy per unit mass, τ is the viscous stress tensor, and q is the heat flow;
[0127] Spatial discretization and linearization
[0128] In the grid generated by the computational domain, for any grid cell S, the steady-state conservation transport equation of a scalar φ can be expressed as:
[0129]
[0130] In the formula
[0131] —loop length vector;
[0132] Γ φ — dissipation coefficient φ;
[0133] — gradient of φ;
[0134] S φ —source term per unit area φ;
[0135] The finite volume method is used to discretize the equation, and the discretized form of the equation on the grid element is:
[0136]
[0137] In the formula
[0138] Nlines — the number of edges surrounding the grid cell S;
[0139] φ f —the value of φ transported through edge f;
[0140] —mass flow through edge f;
[0141] — the length of edge f;
[0142] — The component in the normal direction of edge f;
[0143] S—the area of the grid cell;
[0144] By performing Taylor series expansion on the center value of the grid unit, the flow parameter φ on the boundary line of each grid unit can be obtained: l The second-order upwind format expression of is:
[0145]
[0146] In the formula
[0147] φ—the center value of the grid cell;
[0148] — displacement vector from the centroid of the upstream grid cell to the center of edge f;
[0149] —The gradient of φ is defined as follows:
[0150]
[0151] is the average value of the two grid cells adjacent to the edge f, and the dissipation term is discretized using the central difference format with second-order accuracy;
[0152] The finite volume method is used to discretize the transport equations of turbulent kinetic energy k and turbulent kinetic energy dissipation rate ε. The discrete equations are nonlinear equations containing unknown variables φ and unknown quantities of adjacent grid units. They must be linearized. The linearized form is:
[0153]
[0154] The subscript nb represents the adjacent grid unit, a p and a nb are φ and φ respectively nb The linearization coefficient of
[0155] The discretization of the unsteady equation in the spatial direction is the same as that of the steady equation, and the discretization in the time direction is as follows:
[0156] If the course of a scalar φ changing with time is expressed in differential form as
[0157]
[0158] In the formula, F represents the discrete term in the spatial direction, and the discrete form of the first-order accuracy is
[0159]
[0160] The discrete form of second-order accuracy is
[0161]
[0162] After the time derivative is discretized, F(φ) can be estimated using the φ values at different time levels. According to the different time levels of the φ values used to estimate F(φ), the time direction discretization method is divided into two algorithms: implicit and explicit.
[0163] (1) Implicit format
[0164] In the implicit form, the value of φ at the unknown time layer is used to estimate F(φ), which can be rewritten as
[0165]
[0166] The implicit form of first-order accuracy is
[0167] φ n+1 =φ n +△tF(φ n+1 )
[0168] Nonlinear equations must first be linearized around φn+1 and then solved using an iterative method. The iterative format is
[0169] φ i =φ n +△tF(φ i )
[0170] Similarly, the implicit form of second-order accuracy is
[0171]
[0172] The iteration format is
[0173]
[0174] (2) Display format
[0175] In the explicit format, the unknown time layer is used to estimate F(φ), which can be rewritten as
[0176]
[0177] The φn+1 value of the n+1th time layer can be calculated based on the φn of the existing nth time layer, that is,
[0178] φ n+1 =φ n +△tF(φ n ).
[0179] Furthermore, the thrust verification of the wall-attached flow working state and the switching time verification during the switching process are completed through a single-channel cold air test system. The test system consists of four parts: a high-pressure and large-flow pressure regulating system, a data acquisition and processing system, a jet element test bench dynamic frame and a control console. The high-pressure and large-flow pressure regulating system includes a high-pressure compressor and a high-pressure gas cylinder group, which mainly provide a high-pressure main gas source and control airflow for the jet element; the test bench dynamic frame is used to fix the jet element, and two high-frequency response and high-precision force sensors are installed on the jet element test bench dynamic frame to detect the size of the combined thrust of the jet element in the wall-attached flow working state and the change of the combined thrust during the switching process. The data acquisition and processing system is used to collect the voltage signal transmitted by the force sensor and perform denoising on it; the on and off of the shut-off valve, the switching of the control valve, and the collection and processing of data are controlled by the control console.
[0180] The beneficial effects of the present invention are as follows: a numerical calculation method for the oscillating high-speed jet flow of an underwater solid rocket engine nozzle of the present invention solves the flow field calculation problem of a complex configuration that cannot be solved or has poor effect by a single-domain grid generation method; after the partitioned grid is generated, different regions are allowed to be solved by using different grid systems or different numerical calculation formats; during the solution process after the partitioned grid is generated, at any time, only one subdomain of data is stored in the computer buffer, thereby relaxing the requirements for the computer memory; the use of partitioned grid calculation is suitable for parallel calculation on a multi-processor, has good conservation properties, and is suitable for calculation of internal flow and flow fields with shock waves; it can conveniently describe complex calculation domains, and can use a Cartesian coordinate system to process twisted regions while maintaining an approximate solution with relatively high accuracy; the second-type boundary conditions are as easy to handle as the first-type conditions, and are suitable for calculation of transonic inviscid flows and viscous flows. BRIEF DESCRIPTION OF THE DRAWINGS
[0181] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments made with reference to the following drawings:
[0182] Figure 1 It is a schematic diagram of a jet element test system in a numerical calculation method for high-speed jet flow of an underwater solid rocket engine nozzle swing according to the present invention;
[0183] Figure 2 This is a thrust change waveform diagram of Ps at 80ba in a numerical calculation method for high-speed jet flow of an underwater solid rocket engine nozzle swing according to the present invention;
[0184] Figure 3 This is a thrust variation waveform diagram of Ps of 100ba in a numerical calculation method for high-speed jet flow of an underwater solid rocket engine nozzle swing of the present invention;
[0185] In the figure: 1-air compressor, 2-filter, 3-gas cylinder, 4-pressure reducing valve, 5-pressure gauge, 6-force sensor, 7-supersonic jet element, 8-three-position three-way electromagnetic directional control valve, 9-adjustable flow valve, 10-stop valve. DETAILED DESCRIPTION
[0186] In order to make the technical means, creative features, objectives and effects achieved by the present invention easy to understand, the present invention is further explained below in conjunction with specific implementation methods.
[0187] See also Figure 1 The present invention provides a technical solution: a numerical calculation method for the high-speed jet flow of an underwater solid rocket engine nozzle swing, comprising the following steps:
[0188] Step 1: Establish a mathematical model, using the RANS equation as the control equation, combined with the Realizable k-epsilon turbulence model, near-wall area treatment, closure equation, boundary conditions and initial conditions as the mathematical model for the numerical calculation of supersonic jet elements;
[0189] Step 2: Mesh the mathematical model and fluid calculation domain established in step 1, establish a fluid calculation domain for the underwater solid rocket engine nozzle geometry model obtained in step 1, and perform geometric optimization on the interface between the two fluid calculation domains. Transfer data between the calculation domains to simulate the nozzle swing condition, mesh the flow domain in blocks, and perform mesh independence verification to obtain the optimal number of grids.
[0190] Step 3: Discretization and solution of control equations;
[0191] Step 4: Based on steps 1 to 3, conduct numerical calculation and analysis of the flow characteristics of the underwater solid rocket engine nozzle swinging high-speed jet.
[0192] Further, in step 1, a mathematical model is established including a control equation, a turbulence pattern, a turbulence pattern in a near-wall region, and an improved wall function;
[0193] S1. Control equations Based on the flow characteristics inside the supersonic jet element, the volume force and heat source are ignored, and the wall is assumed to be an adiabatic wall. In the following equations, u i Replace its time mean u i ' is the pulsation value, in the Cartesian coordinate system:
[0194] (1) Continuity equation
[0195]
[0196] (2) Momentum equation
[0197]
[0198] In the formula, Reynolds stress Its expression will be given in the turbulence model, τ ij is the viscous stress tensor, which can be expressed as:
[0199] Strain rate of fluid motion—
[0200] Volume expansion rate of a particle— δ ijis the Kronecker symbol; μ' is the "volumetric viscosity coefficient" or "second viscosity coefficient", which reflects the viscous stress of the fluid deviating from the thermodynamic pressure caused by the volume change. Except for extreme cases such as high temperature and high-frequency sound waves, μ' can be approximately considered to be 0 in general gas motion;
[0201] (3) Energy equation
[0202]
[0203] Where:
[0204] J j' is the diffusion amount of the mixed gas component j'
[0205] in, c p,j' is the constant pressure specific heat of the mixed gas component j', take T ref =298.15K;
[0206] λ is the thermal conductivity, c p is the specific heat at constant pressure, Pr t =0.85
[0207]
[0208] Effective turbulent eddy viscosity coefficient μ eff and the turbulent eddy viscosity coefficient μ t Its expression will be given in the turbulence model;
[0209] S2. In the turbulence mode, there is a strong adverse pressure gradient and flow separation in the flow boundary layer of the supersonic jet element, and under certain conditions, there is a large backflow. The Realizable k-epsilon mode is used to calculate the turbulence. The definitions of turbulent kinetic energy k and turbulent dissipation rate ε are as follows:
[0210]
[0211] (1) Transport equation of turbulent kinetic energy k
[0212]
[0213] Where:
[0214] G k =μ t S 2 , the average strain rate tensor coefficient
[0215] Y M =2ρεM t 2, turbulent Mach number a is the local speed of sound
[0216] (2) Transport equation of turbulent dissipation rate ε
[0217]
[0218] Where:
[0219] And η=Sk / ε
[0220] (3) Turbulent eddy viscosity coefficient μ t
[0221] At high Reynolds numbers, the turbulent eddy viscosity coefficient μ t Expressed as
[0222]
[0223] Where:
[0224] and It is related to the rotating reference surface and is not considered in this calculation;
[0225] A0=4.04,
[0226]
[0227] At low Reynolds numbers, the effective turbulent eddy viscosity coefficient μ eff It can be obtained by the following formula
[0228]
[0229] Where: C υ ≈100
[0230] (4) Reynolds stress
[0231] According to the Boussinesq hypothesis, the Reynolds stress The relationship between the time-averaged velocity gradient is
[0232]
[0233] (5) Model constant
[0234] C2=1.9,σ k =1.0,σ ε =1.2;
[0235] S3. In the turbulence model near the wall, a wall treatment method combining a two-layer wall model with an improved wall function is used: in the two-layer near-wall model, the entire flow area is divided into a viscous influence area and a complete turbulence area according to the size of the turbulence Reynolds number; the turbulence Reynolds number Re y is defined as follows:
[0236] y is the normal distance from the center of the computational grid cell to the wall;
[0237] In the fully turbulent region Use Realizable k-epsilon turbulence model;
[0238] In the near-wall region affected by viscosity The Wolfstein equation model is used. In this model, the transport equation of turbulent kinetic energy k is the same as that of the Realizable k-epsilon turbulence model, and the turbulent eddy viscosity coefficient μ t,2layer and the turbulent dissipation rate ε t,2layer The calculation of is as follows:
[0239] ε t,2layer =k 1.5 / l ε
[0240] In the above formula:
[0241] μ t is the turbulent eddy viscosity coefficient at high Reynolds numbers, and its calculation is shown in the following formula:
[0242] l μ and l ε is the length scale and is calculated as follows:
[0243] l μ =c l y[1-exp(-Re y / A μ )]
[0244] l ε =c l y[1-exp(-Re y / A ε )]
[0245] The constant in the above formula is A μ =70, A ε =2c l
[0246] In order to make the two-layer model defined μ t,2layer It can be defined with the outer layer high Reynolds number μ tSmooth transition, defining an improved turbulent eddy viscosity coefficient μ t,enh , which is defined as follows
[0247] μ t,enh =λ ε μ t +(1-λ ε )μ t,2layer
[0248] Where λ ε is a compromise function, which is used to prevent the convergence of the solution from being blocked when the k-ε solution of the outer layer does not match the solution calculated by the two-layer model. It is defined as follows:
[0249]
[0250] In the above formula, the constant A determines the trade-off function λ ε width; if the variable △Re is given y , to make λ ε The value of A should not exceed 1% of its far-field value. And △Re y The value can be found in Given between 5% and 20% of the value;
[0251] S4. The improved wall function is a hybrid function that organically combines the linear wall law and the logarithmic wall law. For the average velocity, the improved wall function can be expressed as
[0252]
[0253] In the above formula, Where a = 0.01c, b = 5 / c, and
[0254] Taking the derivative of the formula, we get
[0255]
[0256] The improved turbulent wall law can be expressed as
[0257]
[0258] in
[0259]
[0260] This is the position where the slope of the logarithmic law remains constant, and its value is 60 in the calculation;
[0261]
[0262] From the above expressions of α, β and γ, it can be seen that the improved turbulent wall law takes into account the effects of pressure gradient and heat;
[0263] The improved laminar wall law can be expressed as
[0264]
[0265] The improved temperature wall function is expressed as
[0266]
[0267] in, Pr is the molecular Prandtl number;
[0268] S5. Closed-form equations
[0269] In order to make the above equation closed, the following closed equation must be added:
[0270] State equation: f(p,ρ,T)=0
[0271] The density of the flow field is calculated using the complete gas state equation, that is: ρ = p / RT;
[0272] Specific heat at constant pressure:
[0273] R is the universal gas constant, M is the molecular weight of the gas, and f is the number of degrees of freedom of molecular motion. For polyatomic molecules, f=7;
[0274] Thermal conductivity:
[0275] The molecular viscosity coefficient μ is calculated using the Sutherland formula when the temperature is neither too high nor too low, that is,
[0276]
[0277] μ0 is the viscosity coefficient at the reference temperature T0, S is the Sutherland constant;
[0278] For air (210K~1900K), μ0=1.716×10 -5 Pa﹒ s, T0=273K, S=111K;
[0279] When the temperature is low, the Keyes formula is used for calculation, that is,
[0280]
[0281] S6. The supersonic jet element consists of three inlets, two outlets and a wall. Among the three inlets, one is the main gas source inlet and the other two are control flow inlets. According to the flow characteristics of the supersonic jet element and the definition method of boundary conditions in the FLUENT software, the main gas source inlet boundary condition is defined as the pressure inlet boundary condition, the two control flow inlet boundary conditions are defined as the mass inflow boundary conditions, and the two outlet boundary conditions are defined as the pressure outlet boundary conditions. For steady or unsteady flow, when the time-marching algorithm is used to solve, in addition to the boundary conditions, the initial conditions also need to be given. The following is the formulation of each boundary condition and initial condition when calculating the steady or unsteady flow inside the supersonic jet element and the corresponding mathematical treatment;
[0282] (1) The total pressure (P0), total temperature (T0), flow direction, turbulence intensity and hydraulic diameter are given at the pressure inlet boundary conditions. In addition, the static pressure (Ps) is also given to initialize the flow field. From the isentropic relationship of ideal gas, we can get
[0283]
[0284] In the above formula
[0285]
[0286] Where c is the speed of sound, k = cp / cv is the isentropic index,
[0287] The density can be obtained from the ideal gas state equation, that is,
[0288]
[0289] The static temperature Ts in the above formula can be calculated by the following formula
[0290]
[0291] The density, static temperature and velocity scalar values at the pressure inlet can be obtained, and then the velocity component values at the pressure inlet in each direction can be obtained according to the given flow direction;
[0292] (2) Mass inflow boundary conditions
[0293] Gives the mass flow rate Total temperature (T0), static pressure (Ps), flow direction, turbulence intensity and hydraulic directivity
[0294] The mass flow rate is calculated as follows:
[0295]
[0296] In the above formula, vn is the normal velocity and A is the inlet area.
[0297] The density at the inlet is obtained from the ideal gas state equation, that is,
[0298]
[0299] In the above formula, if the flow at the inlet is supersonic, Ps uses a given static pressure value; if the flow at the inlet is subsonic, Ps is obtained by extrapolating the grid center value of the computational grid at the inlet;
[0300] The static temperature in the formula can be calculated by the following formula
[0301]
[0302] In the mass inflow boundary condition, the flux of the corresponding variable at the inlet can be obtained by the normal velocity vn and the inlet value of the variable. The mass flux is ρv n , the flux of turbulent kinetic energy is ρkv n .
[0303] Furthermore, the grid is divided into supersonic jet element grid partitions and boundary layer grids. The entire calculation area is divided into ten subdomains, and the grid of each subdomain is generated separately. Among them, subdomain one adopts an unstructured grid, subdomain eight adopts a hybrid grid in which a structured grid and an unstructured grid coexist, and the remaining eight subdomains adopt a structured grid. When generating calculation grids for each subdomain, the grids at the connection of each subdomain are made completely continuous as much as possible, and no interpolation calculation is required, so as to reduce the amount of information exchange at the regional boundary and make the boundary processing simple. In this way, in the process of capturing cross-region shock waves in the flow field inside the jet element, because the variable transfer at the regional boundary is conserved, the shock wave can be guaranteed to be continuous across regions. According to the flow characteristics near the walls in the supersonic jet element, by giving appropriate h1, m and R values, a boundary layer grid with a density roughly consistent with the change gradient of the flow field parameters can be generated.
[0304] Furthermore, in step three, the control equations are discretized and solved. The flow inside the supersonic jet element is a transonic viscous internal flow, and there is a twisted area. Therefore, the finite volume method is used to discretize the control equations into algebraic equations that can be numerically calculated on the physical space curve grid;
[0305] The finite volume method is described by the control equation in integral form. Therefore, for any two-dimensional control volume S, the control equation is converted into a vector equation in integral form:
[0306]
[0307] In the formula, and They are defined as:
[0308]
[0309] Where l is the loop length of the control volume S, is the velocity vector, u and v are the velocity components in Cartesian coordinates, E is the total energy per unit mass, τ is the viscous stress tensor, and q is the heat flow;
[0310] Spatial discretization and linearization
[0311] In the grid generated by the computational domain, for any grid cell S, the steady-state conservation transport equation of a scalar φ can be expressed as:
[0312]
[0313] In the formula
[0314] —loop length vector;
[0315] Γ φ — dissipation coefficient φ;
[0316] — gradient of φ;
[0317] S φ —source term per unit area φ;
[0318] The finite volume method is used to discretize the equation, and the discretized form of the equation on the grid element is:
[0319]
[0320] In the formula
[0321] Nlines — the number of edges surrounding the grid cell S;
[0322] φ f —the value of φ transported through edge f;
[0323] —mass flow through edge f;
[0324] — the length of edge f;
[0325] — The component in the normal direction of edge f;
[0326] S—the area of the grid cell;
[0327] By performing Taylor series expansion on the center value of the grid unit, the flow parameter φ on the boundary line of each grid unit can be obtained: l The second-order upwind format expression of is:
[0328]
[0329] In the formula
[0330] φ—the center value of the grid cell;
[0331] — displacement vector from the centroid of the upstream grid cell to the center of edge f;
[0332] —The gradient of φ is defined as follows:
[0333]
[0334] is the average value of the two grid cells adjacent to the edge f, and the dissipation term is discretized using the central difference format with second-order accuracy;
[0335] The finite volume method is used to discretize the transport equations of turbulent kinetic energy k and turbulent kinetic energy dissipation rate ε. The discrete equations are nonlinear equations containing unknown variables φ and unknown quantities of adjacent grid units. They must be linearized. The linearized form is:
[0336]
[0337] The subscript nb represents the adjacent grid unit, a p and a nb are φ and φ respectively nb The linearization coefficient of
[0338] The discretization of the unsteady equation in the spatial direction is the same as that of the steady equation, and the discretization in the time direction is as follows:
[0339] If the course of a scalar φ changing with time is expressed in differential form as
[0340]
[0341] In the formula, F represents the discrete term in the spatial direction, and the discrete form of the first-order accuracy is
[0342]
[0343] The discrete form of second-order accuracy is
[0344]
[0345] After the time derivative is discretized, F(φ) can be estimated using the φ values at different time levels. According to the different time levels of the φ values used to estimate F(φ), the time direction discretization method is divided into two algorithms: implicit and explicit.
[0346] (1) Implicit format
[0347] In the implicit form, the value of φ at the unknown time layer is used to estimate F(φ), which can be rewritten as
[0348]
[0349] The implicit form of first-order accuracy is
[0350] φ n+1 =φ n +△tF(φ n+1 )
[0351] Nonlinear equations must first be linearized around φn+1 and then solved using an iterative method. The iterative format is
[0352] φ i =φ n +△tF(φ i )
[0353] Similarly, the implicit form of second-order accuracy is
[0354]
[0355] The iteration format is
[0356]
[0357] (2) Display format
[0358] In the explicit format, the unknown time layer is used to estimate F(φ), which can be rewritten as
[0359]
[0360] The φn+1 value of the n+1th time layer can be calculated based on the φn of the existing nth time layer, that is,
[0361] φ n+1 =φ n +△tF(φ n ).
[0362] The thrust verification of the wall-attached flow working state and the switching time verification during the switching process are completed through a single-channel cold air test system. The test system consists of a high-pressure and large-flow pressure regulating system, a data acquisition and processing system, a jet element test bench frame and a control console. The high-pressure and large-flow pressure regulating system includes a high-pressure compressor and a high-pressure gas cylinder group, which mainly provide a high-pressure main gas source and control airflow for the jet element. The test bench frame is used to fix the jet element. Two high-frequency response and high-precision force sensors are installed on the jet element test bench frame to detect the combined thrust of the jet element in the wall-attached flow working state and the change of the combined thrust during the switching process. The data acquisition and processing system is used to collect the voltage signal from the force sensor and perform denoising on it. The on and off of the stop valve, the switching of the control valve, and the collection and processing of data are controlled by the console.
[0363] The magnitude of the output thrust in the wall-attached flow working state is one of the important indicators to measure the performance of supersonic jet elements. Therefore, this project first calculated the thrust values generated by various main gas source pressures, and then obtained the thrust values corresponding to each main gas source pressure through experiments under the same conditions. Table 1 lists the thrust values corresponding to different main gas source pressures Ps, where test is the test value of the combined thrust generated by the jet element in the X direction. l 、f r They represent the calculated thrust values of the two outlets of the jet element in the X direction, and F is f l and f r When calculating Ps = 80 bar, 90 bar and 100 bar, f l 、f r and F fluctuate periodically. The calculated values in the table are time averages. Error represents the relative error between the calculated value F and the test value Test.
[0364] Table 1 shows the calculated and experimental values of thrust
[0365] Ps(bar) Test(N) <![CDATA[f r (N)]]> <![CDATA[f l (N)]]> F(N) Error(%) 20 44.0 49.65 0.42 49.23 11.89 30 79.9 84.34 1.95 82.40 3.13 40 106.0 108.91 4.43 104.48 -1.43 45 126.3 142.09 4.58 137.51 8.87 50 142.6 164.59 5.66 158.93 11.45 55 161.4 170.27 5.46 164.81 2.11 60 179.3 186.87 6.43 180.44 0.64 70 202.8 220.68 8.55 212.12 4.6 80 239.9 263.05 18.04 245.01 2.1 90 275.4 296.36 21.78 274.58 -0.3 100 313.2 346.18 27.89 318.29 1.63
[0366] It can be seen from Table 1 that the calculated values are in good agreement with the experimental values, especially when the main gas source pressure exceeds 50 bar, the absolute value of the relative error is less than 5%.
[0367] Figure 2 The following is a waveform diagram of thrust variation over time obtained through experiments under the same conditions. The square wave above is the input control signal, the curve below is the combined thrust output measured by the two force sensors, and the curve below is the combined thrust output after noise elimination and smoothing. Figure 2 Ps = 80 bar, Figure 3 Ps = 100 bar, Figure 2 In the middle, the thrust rise time is 1.82ms, and the thrust fall time is 1.28ms; Figure 3In the test, the thrust rise time is 1.18ms and the thrust fall time is 0.85ms.
[0368] From the comparison between the calculated results and the experimental results, it can be seen that the calculated thrust rise time and thrust fall time are close to the experimentally measured data.
[0369] The above shows and describes the basic principles and main features of the present invention and the advantages of the present invention. It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and the present invention can be implemented in other specific forms without departing from the spirit or basic features of the present invention. Therefore, no matter from which point of view, the embodiments should be regarded as exemplary and non-restrictive. The scope of the present invention is defined by the attached claims rather than the above description, and it is intended that all changes falling within the meaning and scope of the equivalent elements of the claims are included in the present invention. Any figure mark in the claims should not be regarded as limiting the claims involved.
[0370] In addition, it should be understood that although the present specification is described according to implementation modes, not every implementation mode contains only one independent technical solution. This description of the specification is only for the sake of clarity. Those skilled in the art should regard the specification as a whole. The technical solutions in each embodiment may also be appropriately combined to form other implementation modes that can be understood by those skilled in the art.
Claims
1. A numerical calculation method for the high-speed jet flow of an underwater solid rocket engine nozzle swing, characterized in that: The steps include: Step 1: Establish a mathematical model, using the RANS equation as the control equation, combined with the Realizable k-epsilon turbulence model, near-wall area treatment, closure equation, boundary conditions and initial conditions as the mathematical model for the numerical calculation of supersonic jet elements; Step 2: Mesh the mathematical model and fluid calculation domain established in step 1, establish a fluid calculation domain for the underwater solid rocket engine nozzle geometry model obtained in step 1, and perform geometric optimization on the interface between the two fluid calculation domains. Transfer data between the calculation domains to simulate the nozzle swing condition, mesh the flow domain in blocks, and perform mesh independence verification to obtain the optimal number of grids. Step 3: Discretization and solution of control equations; Step 4: Based on steps 1 to 3, numerical calculation and analysis of the flow characteristics of the underwater solid rocket engine nozzle swing high-speed jet are carried out; In step 1, a mathematical model including control equations, turbulence patterns, turbulence patterns in the near-wall region, and improved wall functions is established; S1. Control equations Based on the flow characteristics inside the supersonic jet element, the volume force and heat source are ignored, and the wall is assumed to be an adiabatic wall. In the following equations, u i Replace its time mean u i ' is the pulsation value, in the Cartesian coordinate system: (1) Continuity equation (2) Momentum equation In the formula, Reynolds stress- Its expression will be given in the turbulence model, τ ij is the viscous stress tensor, which can be expressed as: Strain rate of fluid motion— Volume expansion rate of a particle— δ ij is the Kronecker symbol; μ' is the "volumetric viscosity coefficient" or "second viscosity coefficient", which reflects the viscous stress of the fluid deviating from the thermodynamic pressure caused by the volume change. Except for extreme cases such as high temperature and high-frequency sound waves, μ' can be approximately considered to be 0 in general gas motion; (3) Energy equation Where: J j' is the diffusion amount of the mixed gas component j' in, c p,j' is the constant pressure specific heat of the mixed gas component j', take T ref =298.15K; λ is the thermal conductivity, c p is the specific heat at constant pressure, Pr t =0.85 Effective turbulent eddy viscosity coefficient μ eff and the turbulent eddy viscosity coefficient μ t Its expression will be given in the turbulence model; S2. In the turbulence mode, there is a strong adverse pressure gradient and flow separation in the flow boundary layer of the supersonic jet element, and under certain conditions, there is a large backflow. The Realizable k-epsilon mode is used to calculate the turbulence. The definitions of turbulent kinetic energy k and turbulent dissipation rate ε are as follows: (1) Transport equation of turbulent kinetic energy k Where: G k =μ t S 2 , the average strain rate tensor coefficient Y M =2ρεM t 2 , turbulent Mach number a is the local speed of sound (2) Transport equation of turbulent dissipation rate ε Where: and η = Sk / ε (3) Turbulent eddy viscosity coefficient μ t At high Reynolds numbers, the turbulent eddy viscosity coefficient μ t Expressed as Where: and It is related to the rotating reference surface and is not considered in this calculation; A0=4.04, At low Reynolds numbers, the effective turbulent eddy viscosity coefficient μ eff It can be obtained by the following formula Where: C υ ≈100 (4) Reynolds stress - By Boussinesq hypothesis, the Reynolds stress - The relationship between the time-averaged velocity gradient is (5) Model constant C2=1.9,σ k =1.0,σ ε =1.2; S3. In the turbulence model near the wall, a wall treatment method combining a two-layer wall model with an improved wall function is used: in the two-layer near-wall model, the entire flow area is divided into a viscous influence area and a complete turbulence area according to the size of the turbulence Reynolds number; the turbulence Reynolds number Re y is defined as follows: y is the normal distance from the center of the computational grid cell to the wall; In the fully turbulent region Use Realizable k-epsilon turbulence model; In the near-wall region affected by viscosity The Wolfstein equation model is used. In this model, the transport equation of turbulent kinetic energy k is the same as that of the Realizable k-epsilon turbulence model, and the turbulent eddy viscosity coefficient μ t,2layer and turbulent dissipation rate The calculation of is as follows: In the above formula: μ t is the turbulent eddy viscosity coefficient at high Reynolds numbers, and its calculation is shown in the following formula; l μ and l ε is the length scale and is calculated as: l μ =c l y[1-exp(-Re y / A μ )] l ε =c l y[1-exp(-Re y / A ε )] The constant in the above formula is A μ =70, A ε =2c l ; In order to make the two-layer model defined μ t,2layer It can be defined with the outer layer high Reynolds number μ t Smooth transition, defining an improved turbulent eddy viscosity coefficient μ t,enh , which is defined as follows: m t,enh =λ ε m t +(1-l ε )m t,2layer Where λ ε is a compromise function, which is used to prevent the convergence of the solution from being blocked when the k-ε solution of the outer layer does not match the solution calculated by the two-layer model. It is defined as follows: In the above formula, the constant A determines the trade-off function λ ε width; if the variable △Re is given y , to make λ ε The value of A should not exceed 1% of its far-field value. And △Re y The value can be found in Given between 5% and 20% of the value; S4. The improved wall function is a hybrid function that organically combines the linear wall law and the logarithmic wall law. For the average velocity, the improved wall function can be expressed as: In the above formula, Where a = 0.01c, b = 5 / c, and Taking the derivative of the formula, we get The improved turbulent wall law can be expressed as in This is the position where the slope of the logarithmic law remains constant, and its value is 60 in the calculation; From the above expressions of α, β and γ, it can be seen that the improved turbulent wall law takes into account the effects of pressure gradient and heat; The improved laminar wall law can be expressed as: The improved temperature wall function is expressed as: in, Pr is the molecular Prandtl number; S5. Closed-form equations In order to make the above equation closed, the following closed equation must be added: State equation: f(p,ρ,T)=0 The density of the flow field is calculated using the complete gas state equation, that is: ρ = p / RT; Specific heat at constant pressure: R is the universal gas constant, M is the molecular weight of the gas, and f is the number of degrees of freedom of molecular motion. For polyatomic molecules, f=7; Thermal conductivity: The molecular viscosity coefficient μ is calculated using the Sutherland formula when the temperature is neither too high nor too low, that is, μ0 is the viscosity coefficient at the reference temperature T0, S is the Sutherland constant; For air (210K~1900K), μ0=1.716×10 -5 Pa﹒ s, T0=273K, S=111K; When the temperature is low, the Keyes formula is used for calculation, that is, S6. The supersonic jet element consists of three inlets, two outlets and a wall. Among the three inlets, one is the main gas source inlet and the other two are control flow inlets. According to the flow characteristics of the supersonic jet element and the definition method of boundary conditions in the FLUENT software, the main gas source inlet boundary condition is defined as the pressure inlet boundary condition, the two control flow inlet boundary conditions are defined as the mass inflow boundary conditions, and the two outlet boundary conditions are defined as the pressure outlet boundary conditions. For steady or unsteady flow, when the time-marching algorithm is used to solve, in addition to the boundary conditions, the initial conditions also need to be given. The following is the formulation of each boundary condition and initial condition when calculating the steady or unsteady flow inside the supersonic jet element and the corresponding mathematical treatment; (1) The total pressure (P0), total temperature (T0), flow direction, turbulence intensity and hydraulic diameter are given at the pressure inlet boundary conditions. In addition, the static pressure (Ps) is also given to initialize the flow field. From the isentropic relationship of ideal gas, we can get In the above formula Where c is the speed of sound, k = cp / cv is the isentropic index, The density can be obtained from the ideal gas state equation, that is, The static temperature Ts in the above formula can be calculated by the following formula The density, static temperature and velocity scalar values at the pressure inlet can be obtained, and then the velocity component values at the pressure inlet in each direction can be obtained according to the given flow direction; (2) Mass inflow boundary conditions Gives the mass flow rate Total temperature (T0), static pressure (Ps), flow direction, turbulence intensity and hydraulic diameter; the mass flow rate is calculated as follows In the above formula, vn is the normal velocity and A is the inlet area. The density at the inlet is obtained from the ideal gas state equation, that is, In the above formula, if the flow at the inlet is supersonic, Ps uses a given static pressure value; if the flow at the inlet is subsonic, Ps is obtained by extrapolating the grid center value of the computational grid at the inlet; The static temperature in the formula can be calculated by the following formula In the mass inflow boundary condition, the flux of the corresponding variable at the inlet can be obtained by the normal velocity vn and the inlet value of the variable. The mass flux is ρv n , the flux of turbulent kinetic energy is ρkv n .
2. The method for numerically calculating the high-speed jet flow of an underwater solid rocket engine nozzle swing according to claim 1, characterized in that: The grid is divided into supersonic jet element grid partition and boundary layer grid. The entire calculation area is divided into ten subdomains, and each subdomain grid is generated separately. Among them, subdomain one adopts unstructured grid, subdomain eight adopts a mixed grid of structured grid and unstructured grid, and the remaining eight subdomains adopt structured grid. When generating calculation grids for each subdomain, try to make the grids at the connection of each subdomain completely continuous, without the need for interpolation calculation, so as to reduce the amount of information exchange at the regional boundary and make the boundary processing simple. In this way, in the process of capturing cross-region shock waves in the flow field inside the jet element, because the variable transfer at the regional boundary is conserved, the shock wave can be guaranteed to be continuous across regions. According to the flow characteristics near the walls in the supersonic jet element, by giving appropriate h1, m and R values, a boundary layer grid with a density roughly consistent with the change gradient of the flow field parameters can be generated.
3. The numerical calculation method for high-speed jet flow of an underwater solid rocket engine nozzle swing according to claim 1, characterized in that: In step three, the control equations are discretized and solved. The flow inside the supersonic jet element is a transonic viscous internal flow, and there is a twisted area. Therefore, the finite volume method is used to discretize the control equations into algebraic equations that can be numerically calculated on the physical space curve grid; The finite volume method is described by the control equation in integral form. Therefore, for any two-dimensional control volume S, the control equation is converted into a vector equation in integral form: In the formula, and They are defined as: Where l is the loop length of the control volume S, is the velocity vector, u and v are the velocity components in Cartesian coordinates, E is the total energy per unit mass, τ is the viscous stress tensor, and q is the heat flow; Spatial discretization and linearization In the grid generated by the computational domain, for any grid cell S, the steady-state conservation transport equation of a scalar φ can be expressed as: In the formula —loop length vector; Γ φ — dissipation coefficient φ; — gradient of φ; S φ —source term per unit area φ; The finite volume method is used to discretize the equation, and the discretized form of the equation on the grid element is: In the formula Nlines — the number of edges surrounding the grid cell S; φ f —the value of φ transported through edge f; —mass flow through edge f; — the length of edge f; — The component in the normal direction of edge f; S—the area of the grid cell; By performing Taylor series expansion on the center value of the grid unit, the flow parameter φ on the boundary line of each grid unit can be obtained: l The second-order upwind format expression of is: In the formula φ—the center value of the grid cell; — displacement vector from the centroid of the upstream grid cell to the center of edge f; —The gradient of φ is defined as follows: is the average value of the two grid cells adjacent to the edge f, and the dissipation term is discretized using the central difference format with second-order accuracy; The finite volume method is used to discretize the transport equations of turbulent kinetic energy k and turbulent kinetic energy dissipation rate ε. The discrete equations are nonlinear equations containing unknown variables φ and unknown quantities of adjacent grid units. They must be linearized. The linearized form is: The subscript nb represents the adjacent grid unit, a p and a nb are φ and φ respectively nb The linearization coefficient of The discretization of the unsteady equation in the spatial direction is the same as that of the steady equation, and the discretization in the time direction is as follows: If the course of a scalar φ changing with time is expressed in differential form as In the formula, F represents the discrete term in the spatial direction, and the discrete form of the first-order accuracy is The discrete form of second-order accuracy is After the time derivative is discretized, F(φ) can be estimated using the φ values at different time levels. According to the different time levels of the φ values used to estimate F(φ), the time direction discretization method is divided into two algorithms: implicit and explicit. (1) Implicit format In the implicit form, the value of φ at the unknown time layer is used to estimate F(φ), which can be rewritten as The implicit form of first-order accuracy is f n+1 =φ n +△tF(φ n+1 ) Nonlinear equations must first be linearized around φn+1 and then solved using an iterative method. The iterative format is f i =φ n +△tF(φ i ) Similarly, the implicit form of second-order accuracy is The iteration format is (2) Display format In the explicit format, the unknown time layer is used to estimate F(φ), which can be rewritten as The φn+1 value of the n+1th time layer can be calculated based on the φn of the existing nth time layer, that is, f n+1 =φ n +△tF(φ n )。 4. The method for numerically calculating the high-speed jet flow of an underwater solid rocket engine nozzle swing according to claim 1, characterized in that: The thrust verification of the wall-attached flow working state and the switching time verification during the switching process are completed through a single-channel cold air test system. The test system consists of four parts: a high-pressure and large-flow pressure regulating system, a data acquisition and processing system, a jet element test bench dynamic frame and a control console. The high-pressure and large-flow pressure regulating system includes a high-pressure compressor and a high-pressure gas cylinder group, which mainly provide a high-pressure main gas source and control airflow for the jet element; the test bench dynamic frame is used to fix the jet element, and two high-frequency response and high-precision force sensors are installed on the jet element test bench dynamic frame to detect the size of the combined thrust of the jet element in the wall-attached flow working state and the change of the combined thrust during the switching process. The data acquisition and processing system is used to collect the voltage signal from the force sensor and perform denoising on it; the on and off of the stop valve, the switching of the control valve, and the collection and processing of data are controlled by the console.
