Optimization design method for geometric parameters of wind tunnel diffusion section of coupled BIM and hybrid RANS-LES model

Abstract

The invention relates to the technical field of wind tunnel diffusion section structure optimization, and discloses a wind tunnel diffusion section geometric parameter optimization design method coupling BIM and a hybrid RANS-LES model, and the method comprises the steps: building a hybrid RANS-LES model for wind tunnel diffusion section flow field simulation, employing the RANS model to simulate a near-wall region, and employing a large-eddy simulation model (LES) to simulate other regions; a multi-parameter BIM model of the wind tunnel diffusion section special-shaped structure is established based on an NURBS curved surface modeling method, and parametric modeling and automatic updating of the wind tunnel diffusion section special-shaped structure are achieved; a wind tunnel diffusion section BIM model is divided into computational grids for CFD numerical simulation, then diffusion section geometric parameters are optimized according to a simulation result, automatic updating of the wind tunnel diffusion section model is achieved through BIM parametric modeling, and then bidirectional coupling of the BIM model and CFD numerical simulation is achieved.

Description

Technical Field

[0001] This application relates to the technical field of wind tunnel diffuser structure optimization. Specifically, it involves a geometric parameter optimization design method for wind tunnel diffusers that couples BIM and the hybrid RANS-LES model. Background Art

[0002] The wind tunnel diffuser is an important structure that ensures that the airflow does not diffuse and separate when passing through, and reduces the airflow pulsation to ensure a uniform velocity distribution at the inlet of the downstream stable section. The design of the diffuser geometric parameters is a prerequisite for ensuring the function of the wind tunnel diffuser. Using the Computational Fluid Dynamics (CFD) method to simulate the diffuser flow field has become a necessary means for optimizing the design of the diffuser geometric parameters.

[0003] However, existing studies mostly use the Reynolds Averaged Navier-Stokes (RANS) model to simulate the diffuser flow field. The calculation results are time-averaged physical quantities that omit the instantaneous characteristics, and it is difficult to capture the turbulent transient pulsation in the simulation. In addition, the special-shaped structure of the square-round transition section has problems of low 3D modeling accuracy and low efficiency, which in turn affect the flow field simulation efficiency and result accuracy.

[0004] Therefore, this study established a hybrid RANS-LES model for simulating the wind tunnel diffuser flow field. The RANS model is applied to simulate the near-wall region, and the Large Eddy Simulation (LES) model is applied to simulate the rest of the region, so as to realize the simulation of the anisotropic vortex structure in the wind tunnel diffuser turbulence. On this basis, a multi-parameter BIM model of the special-shaped structure of the wind tunnel diffuser is established based on the NURBS surface modeling method to realize the parametric update of the special-shaped structure of the wind tunnel diffuser, and the geometric size optimization is realized by coupling the diffuser flow field simulation. Summary of the Invention

[0005] This application aims to solve at least one of the technical problems existing in the prior art. For this purpose, this application proposes a geometric parameter optimization design method for wind tunnel diffusers that couples BIM and the hybrid RANS-LES model, including:

[0006] S1. Establish a hybrid RANS-LES model for CFD simulation of the wind tunnel diffuser based on the Realizable k-ε turbulence model. Among them, the LES method is used for the mainstream region dominated by large-scale vortices, and the RANS method is used for the near-wall region dominated by small-scale vortices, taking into account both calculation accuracy and calculation efficiency;

[0007] S2. Establish a multi-parameter BIM model of the special-shaped structure of the diffuser section using the parametric modeling method based on NURBS surfaces. The size control parameters are the circular end diameter D, the square end side length A, and the diffuser section length L.

[0008] S3. Conduct two-way coupling between the BIM model and CFD numerical simulation. Divide the BIM parametric modeling results of the wind tunnel diffuser section into computational grids and use them as the computational model to input into the subsequent CFD numerical simulation of the flow field. Subsequently, guide the size optimization of the diffuser section according to the CFD numerical simulation results of the flow field, and achieve automatic update of the wind tunnel diffuser section through BIM parametric modeling.

[0009] For a geometric parameter optimization design method of a wind tunnel diffuser section coupling BIM and a hybrid RANS-LES model according to an embodiment of the present application, the beneficial effects are as follows:

[0010] Use the hybrid RANS-LES model to conduct flow field simulation of the diffuser section, improving the deficiency that the simulation results of existing research using only the RANS model are difficult to capture the turbulent transient pulsation; at the same time, establish a multi-parameter BIM model of the special-shaped structure of the wind tunnel diffuser section based on the NURBS surface modeling method, and conduct two-way coupling between the BIM model and CFD numerical simulation to achieve size optimization and parametric update of the special-shaped structure of the wind tunnel diffuser section. Description of the Drawings

[0011] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for the embodiments will be briefly introduced below. It should be understood that the following drawings only show some embodiments of the present application, and therefore should not be regarded as limiting the scope. For those of ordinary skill in the art, other related drawings can be obtained based on these drawings without creative efforts.

[0012] Figure 1 are the characteristics of the NURBS curve according to an embodiment of the present application;

[0013] Figure 2 is the automatic update of the NURBS curve according to an embodiment of the present application;

[0014] Figure 3 is the NURBS tensor product construction surface according to an embodiment of the present application;

[0015] Figure 4 is the BIM parametric modeling process of the wind tunnel diffuser section according to an embodiment of the present application;

[0016] Figure 5 is the comparison of the simulation results of the flow velocity distribution of the multi-scheme wind tunnel diffuser section according to an embodiment of the present application. Detailed Embodiments

[0017] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts fall within the scope of protection of this application.

[0018] The following details a method for optimizing the geometric parameters of the wind tunnel diffuser section that couples BIM and the hybrid RANS-LES model through specific embodiments.

[0019] S1 Multi-parameter BIM model of the special-shaped structure of the wind tunnel diffuser section based on NURBS

[0020] The three-dimensional modeling of the wind tunnel diffuser section is the basis for theoretical analysis and construction control. The wind tunnel diffuser section is a special-shaped structure with a gradual change from square to round. The conventional grid-based modeling method is a discretized modeling method, and the established model can only ensure first-order continuity of the boundary within a certain accuracy requirement range, not completely smooth mathematically. At the same time, when modifying the geometric dimensions of the wind tunnel diffuser section, the update efficiency of the special-shaped structure model is relatively low. Therefore, a multi-parameter BIM model of the special-shaped structure of the diffuser section is established based on NURBS to achieve a smooth surface mathematical expression and parametric BIM modeling of the model. A p-order NURBS curve is defined as shown in Equation (1):

[0021]

[0022] In Equation (1): C(t) is a free curve with parameter t; {P i} is a set of control points, and the set contains n + 1 control points; {ω i} is a set of weight factors; {N i,p (t)} is a p-order B-spline basis function, and the basis function is determined by a periodic (C(a) = C(b)) or non-periodic (C(a) ≠ C(b)), uniform (ti+1 - ti = ti - ti-1) or non-uniform ((ti+1 - ti ≠ ti - ti-1)) knot vector (KnotVector) U, where the knot vector U of a p-order NURBS curve with m (m = n + p + 1) knots is expressed as Equation (2), and the corresponding knot set of the knot vector U is {Qk}.

[0023]

[0024] For any NURBS curve, it is possible to move the control point P i or change the weight factor ω iThe modification of the local geometry of the curve is achieved in a certain way, and changing the division method of the parameter interval (i.e., the parameter values of the nodes Qk) will not have any impact on the curve geometry, such as Figure 1 (a, b), the curve C(t) is a single-valued vector function of the parameter t. Based on this property, assuming that all weight factors are equal to 1, the parameter interval is (0, 1), the coordinate constraints are used as the node set {Qk}, and corresponding parameter values are specified for each Qk (k = 0, 1,..., n) d is the total length of the polyline formed by {Qk}), the node vector U and the basis function {N i,p (t)} can be determined, and thus the NURBS curve based on the interpolation points can be obtained, as shown in Equation (3), and the curve can be automatically updated by adding a new coordinate constraint Qnew, such as Figure 1 , Figure 2 .

[0025]

[0026] Model the surface in the form of a tensor product topology of longitudinal and transverse NURBS curves, as shown in Equation (4). Since the NURBS curve is a single-valued vector function of the parameter t, the NURBS surface is a single-valued vector function of the parameters u and v, realizing parametric control in the u and v directions, such as Figure 3 shown.

[0027]

[0028] In Equation (4), S(u, v) is the NURBS surface with the longitudinal parameter u and the transverse parameter v as control parameters; {P i，j} is the control point set forming the control grids in two directions; {N i,p (u)} and {N j,q (v)} are the non-rational B-spline basis functions of order p and q defined in the u and v directions respectively; n and m are the total number of divided segments of the parameter intervals in the u and v directions respectively.

[0029] Subsequently, perform BIM parametric modeling on the special-shaped structure of the wind tunnel diffuser section based on the NURBS surface. The diffuser section is a square-round gradual change section, consisting of four triangular planes and four blending surfaces. The main control parameters are the side length A (m) of the square end, the diameter D (m) of the circular end, and the total length L (m) of the diffuser section. The overall idea of the parametric modeling program is as follows: First, divide the circular end into four equal parts, and establish triangular planes with the four equal division points and the corresponding vertexes of the square end side; Subsequently, use the hypotenuses of each triangular plane and the adjacent triangular planes as the sweeping paths, and the quarter arc of the corresponding circular end as the sectional line to perform double-rail sweeping operations to obtain the square-round gradual change section model. The wind tunnel diffuser section model can be automatically updated by modifying the parameters A, D, and L, and the model is a mathematically smooth surface without fitting errors, ensuring the accuracy of the model, such as Figure 4 shown.

[0030] Finally, a two-way coupling between the BIM model and CFD numerical simulation is carried out. The parametric modeling results of the wind tunnel diffuser in BIM are divided into computational grids and used as the computational model to be input into the subsequent CFD numerical simulation of the flow field. Subsequently, the size optimization of the diffuser is guided according to the results of the CFD numerical simulation of the flow field, and the automatic update of the wind tunnel diffuser is realized through parametric modeling in BIM.

[0031] S2 Flow field control equations

[0032] The basic idea of the hybrid RANS-LES is to use the RANS turbulence model for simulation in the near-wall flow region where small-scale vortices dominate the motion, and use the LES model for direct calculation in the main region where large-scale vortices dominate the motion. Therefore, the basic control equations of the flow field should include two parts: the basic control equations of RANS and the basic equations of LES. The mass and momentum conservation control equations of RANS are as shown in Eqs. (5,6):

[0033]

[0034] In Eqs. (5,6), ~ represents the variables shared between RANS and LES; ρ is the density, kg / m3; t is the computational time, s; u i is the velocity of the fluid in the x, y, and z directions, kg / m3; p is the pressure, Pa; v is the kinematic viscosity coefficient, m2 / s; F is the momentum source term, m2 / s.

[0035] The LES method filters out small-scale vortices through a filtering filter and directly solves the control equations for large-scale vortices. The mathematical expression of the filtering filter is as shown in (7):

[0036]

[0037] In Eq. (7), is the variable after large-scale filtering, G(x, y) is the Gaussian filtering function, and D is the fluid region.

[0038] After the filtering process, the basic control equations of LES can be obtained as shown in Eqs. (8,9):

[0039]

[0040] In Eq. (9), σ ij is the viscous stress tensor; τ ij is the sub-grid Reynolds stress obtained by filtering small-scale fluctuations based on the local equilibrium assumption.

[0041] S3 Hybrid RANS-LES model

[0042] A hybrid RANS-LES model is established based on the Realizable k-ε turbulence model for numerical simulation of the flow field in the diffuser section of a wind tunnel. In the Realizable k-ε turbulence model, the governing equations for the turbulent kinetic energy k and the dissipation rate ε are as shown in Eqs. (10,11):

[0043]

[0044] In Eqs. (10,11), d is the spatial scale of the computational model; μ t is the turbulent viscosity coefficient, μ t =ρC μ k 2 / ε, N·m / s; G k is the turbulent kinetic energy generated by the mean velocity gradient, m 2 / s 2 ; σk and σε are the Prandtl constants related to the turbulent kinetic energy k and the dissipation rate ε, with values of 1.0 and 1.2 respectively; C1 and C2 are empirical constants, the calculation method of C1 is as shown in Eq. (12), and C2 has a value of 2.0; S is the turbulent strain rate tensor, and its calculation method is as shown in Eq. (14);

[0045]

[0046] The hybrid RANS-LES model is a computational method that combines the Reynolds-averaged Navier-Stokes method and the large eddy simulation method. The model spatial scale d is obtained by mixing the spatial scale dRANS of the Reynolds-averaged model and the spatial scale dLES of the large eddy model. When the spatial scale dRANS of the Reynolds-averaged model dominates, the Realizable k-ε turbulence model is used for calculation in the region near the wall; when the spatial scale dLES of the large eddy model dominates, the filtered LES governing equations are directly solved in the flow core region far from the wall. The calculation method of d is as shown in Eqs. (15 - 18):

[0047]

[0048] d LES =C DES Δ (17)

[0049] Δ=max(Δx, Δy, Δz)(18)

[0050] In Eqs. (15 - 18), Δ is the maximum computational grid size, m; Δx, Δy, and Δz are the computational grid sizes in the x, y, and z directions respectively, m; C DES is the computational model calibration constant, with a value of 0.61.

[0051] The hybrid RANS-LES model combines the characteristic scales of turbulence defined by the grid size through equations (15-18) to achieve the hybridization of different solution models. When k 2 / 3 / ε ≤ C DES Δ, the spatial scale dRANS of the RANS model dominates in the region near the wall. The calculation method of the turbulent viscosity coefficient μ t of the RANS-LES hybrid model is the same as that of the Realizable k-ε turbulence model, which is μ t = ρC μ k 2 / ε; when k 2 / 3 / ε > CDESΔ, the spatial scale dLES of the LES model dominates in the flow core region far from the wall. The calculation method of the turbulent viscosity coefficient μt of the hybrid RANS-LES model is the same as that of the LES model, which is μ t = ρC μ k 0.5 d.

[0052] Example of geometric parameter optimization of the wind tunnel diffuser section by coupling BIM with the hybrid RANS-LES model

[0053] The hybrid RANS-LES model is used to numerically simulate the diffuser section under different combinations of size parameters. Based on the previous BIM model of the diffuser section, the main size parameters are the round-end diameter D, the square-end side length A, and the diffuser section length L. Taking three combination schemes of the above size parameters as examples, a hybrid RANS-LES numerical simulation study is carried out on the ventilation process of the diffuser section, and the contour map of the longitudinal section wind speed distribution of the diffuser section along the central axis after ventilation stability is obtained, as shown in Figure 5 . The three combination schemes of size parameters are (a) D = 9m, A = 11m, L = 33m; (b) D = 10m, A = 14m, L = 40m; (c) D = 8m, A = 12m, L = 25m.

[0054] The wind tunnel diffuser section is an important structure to ensure that the air flow does not diffuse and separate when passing through, and to reduce the air flow pulsation to ensure the uniform velocity distribution at the inlet of the downstream stable section. The smaller the diffusion separation part, the higher the flow field quality of the diffuser section, indicating that the size parameter design of the diffuser section is reasonable. From Figure 5It can be seen that in the three combinations of size parameters, the flow field of scheme (a) shows the diffusion separation in the least area, the outlet flow velocity distribution is relatively uniform, reaching 12 - 14 m / s, and the available flow field accounts for 81.4% of the entire cross-section. This is because the size difference between the round end and the square end is small and the length of the diffusion section is long, so the flow field in the diffusion section can transition smoothly and fully. However, large diffusion separations occur in both schemes (b) and (c), and there are large gradient changes in the outlet flow velocity along the vertical direction, resulting in poor flow field quality. The outlet flow velocity of scheme (b) is 12 - 14 m / s, and the available flow field only accounts for 57.9% of the entire cross-section; the outlet flow velocity of scheme (c) is only 10 - 12 m / s, and the available flow field accounts for 65.2% of the entire cross-section. Therefore, by comprehensively comparing and selecting to optimize the size parameters of the diffusion section, scheme (a) with D = 9.5 m, A = 11 m, and L = 33 m is finally selected as the final design result.

[0055] It should be noted that Figure 1 in, (a) and (b) reflect that the division method of the parameter interval does not affect the geometric shape of the curve, and (c) and (d) reflect the characteristics of the single-valued vector function of the NURBS curve.

[0056] The above embodiments are only used to illustrate the specific embodiments of the present invention and are not limited thereto. For those skilled in the art, according to the idea of the present invention, various similar deformations and transformations can be made, and these deformations and transformations should all be regarded as the protection scope of the present invention.

Claims

1. A wind tunnel diffuser geometric parameter optimization design method coupling BIM and hybrid RANS-LES model, characterized by: The following steps are involved: S1. A hybrid RANS-LES model for CFD simulation of the diffuser section of a wind tunnel is established, in which the LES method is used for the mainstream area dominated by large-scale vortices, and the RANS method is used for the near-wall area dominated by small-scale vortices, taking into account both computational accuracy and efficiency; S2. Establish a multi-parameter BIM model of the special-shaped structure of the wind tunnel diffusion section; S3. Carry out bidirectional coupling between BIM model and CFD numerical simulation.

2. The geometric parameter optimization design method of the wind tunnel diffuser section coupling BIM and hybrid RANS-LES model according to claim 1 is characterized in that: The hybrid RANS-LES model for CFD simulation of the wind tunnel diffusion section in S1 is established based on the Realizable k-ε turbulence model.

3. The geometric parameter optimization design method of the wind tunnel diffuser section coupling BIM and hybrid RANS-LES model according to claim 1 is characterized in that: In the hybrid RANS-LES model of the CFD simulation of the diffuser section of the wind tunnel in S1, the basic control equation of RANS is:

4. The method for optimizing geometric parameters of a wind tunnel diffuser section by coupling a BIM and a hybrid RANS-LES model according to claim 1, characterized in that: In the hybrid RANS-LES model of the CFD simulation of the diffuser section of the wind tunnel in S1, the basic control equation of LES is:

5. The geometric parameter optimization design method of the wind tunnel diffuser section coupling BIM and hybrid RANS-LES model as claimed in claim 2 is characterized in that: In the hybrid RANS-LES model based on the Realizable k-ε turbulence model, the governing equations of turbulent kinetic energy k and dissipation rate ε are:

6. The geometric parameter optimization design method of the wind tunnel diffuser section coupling BIM and hybrid RANS-LES model as claimed in claim 5 is characterized in that: In the hybrid RANS-LES model, the model space scale d is calculated as: d LES =C DES Δ and Δ = max(Δx, Δy, Δz).

7. The method for optimizing geometric parameters of a wind tunnel diffuser section by coupling a BIM and a hybrid RANS-LES model according to claim 1, characterized in that: The multi-parameter BIM model of the special-shaped structure of the diffuser section of the wind tunnel in S2 is established based on the NURBS surface, and the size control parameters are the round end diameter D, the square end side length A and the diffuser section length L.

8. The method for optimizing geometric parameters of a wind tunnel diffuser section by coupling a BIM and a hybrid RANS-LES model as claimed in claim 7, characterized in that: When establishing a multi-parameter BIM model of the special-shaped structure of the wind tunnel diffuser based on the NURBS surface, the NURBS curve is defined as:

9. The method for optimizing geometric parameters of a wind tunnel diffuser section by coupling a BIM and a hybrid RANS-LES model according to claim 1, characterized in that: When establishing a multi-parameter BIM model of the special-shaped structure of the wind tunnel diffuser based on the NURBS surface, the NURBS surface is defined as:

10. The method for optimizing geometric parameters of a wind tunnel diffuser section by coupling BIM and hybrid RANS-LES model according to claim 1, characterized in that: The bidirectional coupling between the BIM model and the CFD numerical simulation in S3 includes dividing the BIM parametric modeling results of the wind tunnel diffuser into a computational grid and inputting the grid into the subsequent flow field CFD numerical simulation as a computational model, guiding the size optimization of the diffuser according to the flow field CFD numerical simulation results, and realizing automatic updating of the wind tunnel diffuser through BIM parametric modeling.

Images