Topological optimization design method for radiator flow channel structure based on body-fitted grid

Abstract

The invention belongs to the related technical field of radiator flow channel optimization design, and discloses a radiator flow channel structure topology optimization design method based on a body-fitted grid, which comprises the following steps: (1) on the basis of an initial level set function of a radiator to be optimized, constructing the body-fitted grid fitted with a fluid-solid interface according to a zero isoline of the level set function; (2) carrying out material mapping of a binary structure on the fluid-solid system according to the fluid-solid boundary, and solving a fluid-heat coupling control equation of the geometric model to be optimized by utilizing the obtained mapping relation so as to obtain the flowing and radiating states of thermal fluid in the radiator; and (3) solving the constructed topological optimization model until convergence, and outputting a corresponding level set function during convergence, so that a structure corresponding to the obtained level set function is an optimized runner structure. According to the method, automatic encryption can be realized in a structure boundary region, and sparse distribution of grids is kept in a region far away from the boundary, so that a clearer and smoother structure boundary is obtained.

Description

Technical Field

[0001] The present invention belongs to the technical field related to the optimization design of radiator flow channels, and more specifically, relates to a method for topologically optimizing the structure of a radiator flow channel based on body-fitted grids. Background Art

[0002] With the rapid development of fields such as aerospace, the electronics industry, and equipment manufacturing, the requirements for the heat dissipation performance of equipment are constantly increasing. For example, regenerative cooling channels in aeroengines, finned radiators in electronic chips, water-cooled heat dissipation channels in lithium-ion batteries, and conformal cooling molds in injection molding all require efficient heat dissipation structure designs to ensure the stable operation and long life of the system. Traditional radiator structures based on parametric design, due to limited design parameters, effectively reduce the flow resistance of the coolant while achieving efficient heat dissipation, thus restricting the improvement of the overall thermal management performance. In contrast, structural topology optimization technology has gradually become the main optimization tool for radiator structure design due to its greater design freedom and strong multi-objective design capabilities.

[0003] Currently, the topological optimization methods for radiators mainly include two types: the variable density method and the level set method. Among them, the variable density method is widely used because of its relatively simple implementation. However, such methods often have problems with jagged boundaries and cannot completely eliminate the intermediate density cells in the design results, making it impossible to clearly express the structural boundaries, and thus requiring additional smoothing processing to meet the requirements of machining and manufacturing. In contrast, the level set method can more accurately describe geometric boundaries and has strong mathematical rigor, enabling the natural smooth evolution of structural boundaries. However, currently, the level set method usually relies on fixed regular grids, and the level set function cuts the grids in the boundary region, which limits its boundary description ability. Summary of the Invention

[0004] Aiming at the above defects or improvement requirements of the prior art, the present invention provides a method for topologically optimizing the structure of a radiator flow channel based on body-fitted grids, which aims to solve the problem that the boundary description ability of the existing level set function used for topological optimization of heat dissipation structures is limited due to cutting grids in the boundary region.

[0005] To achieve the above object, according to one aspect of the present invention, there is provided a method for topologically optimizing the structure of a radiator flow channel based on body-fitted grids, the method comprising the following steps:

[0006] (1) Based on the initial level set function of the radiator to be optimized, construct a body-fitted grid that fits the fluid-solid interface according to the zero isocontour of the level set function;

[0007] (2) Map the fluid-structure system into a binary structure according to the fluid-structure boundary, and use the obtained mapping relationship to solve the fluid-thermal coupling control equation of the geometric model to be optimized, so as to obtain the internal thermal fluid flow and heat dissipation state of the radiator;

[0008] (3) Considering the multi-objectives of maximizing structural heat dissipation and minimizing fluid dissipated energy, use the adaptive weighted method to establish the objective function, with the maximum volume fraction and the maximum fluid dissipated energy as inequality constraints, the fluid-thermal coupling control equation as an equality constraint, and the level set function as the design variable to construct a topology optimization model based on body-fitted grids;

[0009] (4) Solve the constructed topology optimization model until convergence, output the corresponding level set function at convergence, and the structure corresponding to the obtained level set function is the optimized flow channel structure.

[0010] Further, considering the multi-objectives of maximizing structural heat dissipation and minimizing fluid dissipated energy, use the adaptive weighted method to establish the objective function, with the maximum volume fraction and the maximum fluid dissipated energy as inequality constraints, the fluid-thermal coupling control equation as an equality constraint, and the level set function as the design variable to construct a topology optimization model.

[0011] Further, the initial level set function φ describes the fluid phase and solid phase of the geometric model respectively. By updating the level set function value, the material distribution of the fluid-structure system is changed, so as to obtain the flow channel structure expressed by the level set function. Among them, the mapping relationship between the level set function and the material distribution of the fluid-structure system in the geometric model is:

[0012]

[0013] In the formula, D represents the model calculation domain, Ω represents the fluid domain, and the complementary domain D\Ω represents the solid domain, represents the fluid-structure boundary.

[0014] Further, use the body-fitted grid technology to adaptively reconstruct the grid of the geometric model. According to the zero isocontour of the level set function, the grid of the geometric model is divided into different sub-grids with explicit boundaries. The divided sub-grids represent the solid phase and fluid phase of the geometric model respectively, realizing the explicit expression of the fluid-structure interface at the grid level.

[0015] Further, the fluid-thermal coupling control equation is:

[0016]

[0017] In the formula, Re is the Reynolds number, Pr is the Prandtl number, ▽ is the dimensionless gradient operator, the state variable u is the dimensionless velocity, p is the dimensionless pressure, T is the dimensionless temperature, the material parameter α is the inverse permeability, and β is the temperature difference heat transfer coefficient.

[0018] Furthermore, under the background of the reconstructed body-fitted grid, the level set function φ is mapped to the characteristic function χ according to the explicit fluid-structure boundary, φ and a material mapping of the 1 / 0 binary structure is performed on the fluid-structure system. The expression of the obtained mapping relationship is:

[0019]

[0020] Based on the obtained mapping relationship, a rational approximation model of material properties is adopted for material interpolation. The formula corresponding to material interpolation is:

[0021]

[0022] In the formula, q a , q b are interpolation penalty factors; α max is the maximum inverse permeability; β max is the maximum heat transfer coefficient of temperature difference.

[0023] Furthermore, the reaction-diffusion equation (RDE) is used to update the level set function. This equation is divided into two parts: the reaction term and the diffusion term. The reaction term contains the sensitivity of the objective function with respect to the design variables, enabling automatic opening of holes during the topology evolution process. The diffusion term ensures the regularity of the level set function and avoids the traditional level set re-regularization process. The equation is as follows:

[0024]

[0025] Among them, φ is the level set function, t is the virtual evolution time, K is the normalization scale coefficient, is the optimization sensitivity, τ is the diffusion coefficient, and n is the unit normal vector from the boundary to the outside. The optimization sensitivity is obtained by the continuous adjoint method, and the magnitude of the diffusion coefficient can adjust the geometric complexity of the topological configuration.

[0026] Furthermore, the optimization formulation of the topology optimization model is:

[0027]

[0028] Among them, G1 is the volume constraint, V max is the upper limit of the allowable volume fraction of the fluid domain, G2 is the dissipated energy constraint, is the upper limit of the allowable fluid dissipated energy.

[0029] Generally speaking, compared with the prior art by the above technical solutions conceived by the present invention, the method for topological optimization design of the radiator flow channel structure based on the body-fitted grid provided by the present invention mainly has the following

[0030] beneficial effects:

[0031] 1. The present invention uses a body-fitted grid technology to adaptively reconstruct the model grid. According to the zero isosurface of the level set function, the model grid is divided into fluid-phase and solid-phase sub-grids with explicit boundaries. This method can achieve automatic encryption in the structural boundary region, while maintaining a sparse grid distribution in the region far from the boundary, thereby obtaining a clearer and smoother structural boundary, which is suitable for direct machining and manufacturing. At the same time, this technology can also reduce the computational cost to a certain extent and improve the computational efficiency.

[0032] 2. The present invention models the fluid-structure interaction system through a material mapping method of a binary structure, without setting any continuous material interpolation format to penalize the intermediate density cells, which improves the solution accuracy.

[0033] 3. The present invention comprehensively considers two mutually restrictive optimization objectives of heat dissipation and flow resistance, and adopts an adaptive update strategy of a numerical matching coefficient to avoid the influence of the order-of-magnitude difference between the two objective functions on the optimization process. The flow channel structure under different target weight distributions can be obtained to guide the radiator design in specific scenarios, which improves the practicality in the engineering field.

[0034] 4. The present invention uses the continuous adjoint sensitivity method to deduce the sensitivity and updates the level set function by solving the reaction-diffusion equation. Compared with the traditional level set method, it can automatically open holes, and the introduction of the diffusion term ensures the regularity of the level set function, avoiding the re-regularization process of the traditional level set, ensuring the accuracy of the sensitivity solution and the stability of the iterative process. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] Figure 1 In (a) and (b) of, they are respectively a schematic diagram of the level set isosurface representing the structural boundary and a schematic diagram of the body-fitted grid showing and encrypting the structural boundary;

[0036] Figure 2 is the flow chart of the topology optimization;

[0037] Figure 3 is the schematic diagram of the design space and boundary conditions;

[0038] Figure 4 is the iteration curve of the objective and constraint functions;

[0039] Figure 5 is the schematic diagram of the result obtained from the topology optimization design of the heat transfer and flow coupling of the radiator flow channel. DETAILED DESCRIPTION OF THE INVENTION

[0040] To make the objectives, technical solutions and advantages of the present invention more clear and understandable, the present invention will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention. In addition, the technical features involved in the various embodiments of the present invention described below can be combined with each other as long as they do not conflict with each other.

[0041] Please refer to Figure 1 , the present invention provides a topology optimization design method for the radiator flow channel structure based on body-fitted grids. The optimization design method is a level set topology optimization design method for conjugate heat transfer channels, which avoids the material interpolation error caused by the level set cutting the fixed grid in the traditional method, obtains a clear and smooth fluid-solid boundary, designs a flow channel structure that meets the heat dissipation and flow resistance requirements, and improves the radiator performance. At the same time, the optimization design method takes into account both heat dissipation and flow resistance, can obtain a clear and smooth optimization result, improves the calculation efficiency and solution accuracy, and improves the practicality in the engineering field.

[0042] The optimization design method mainly includes the following steps:

[0043] Step 1, based on the initial level set function of the radiator to be optimized, construct a body-fitted grid that fits the fluid-solid interface according to the zero isosurface of the level set function.

[0044] Construct the geometric model of the radiator to be optimized, determine the design domain, non-design domain, and flow field and temperature boundary conditions; construct the initial level set function on the geometric model, and construct a body-fitted grid that fits the fluid-solid interface according to the zero isosurface of the level set function.

[0045] The initial level set function φ describes the fluid phase and solid phase of the geometric model respectively. By updating the level set function value, the material distribution of the fluid-solid system is changed, so as to obtain a flow channel structure expressed by the level set function. Among them, the mapping relationship between the level set function and the material distribution of the fluid-solid system in the geometric model is:

[0046]

[0047] In the formula, D represents the model calculation domain, Ω represents the fluid domain, and the complementary domain D\Ω represents the solid domain, represents the fluid-solid boundary.

[0048] The specific construction steps of the body-fitted grid are as follows: adopt the body-fitted grid technology to adaptively reconstruct the grid of the geometric model, divide the grid of the geometric model into different sub-grids with explicit boundaries according to the zero isosurface of the level set function. The divided sub-grids represent the solid phase and fluid phase of the geometric model respectively, realizing the explicit expression of the fluid-solid interface at the grid level, and automatically encrypting the grid in the boundary area to obtain a clear and smooth high-resolution flow channel structure boundary, as Figure 1As shown. The body-fitted grid technology allows the independent setting of relevant parameters during the optimization process, including the maximum grid size h max , the minimum grid size h min , the grid change gradient hgrad, and the Hausdorff distance hausd; where the grid change gradient represents the ratio of the lengths of adjacent grid edges, and the Hausdorff distance represents the approximation accuracy between the grid model and the geometric model. The boundary grid can be refined by reducing the Hausdorff distance.

[0049] Step 2: Perform a binary structure material mapping on the fluid-structure system according to the fluid-structure boundary, and use the obtained mapping relationship to solve the fluid-thermal coupling control equation of the geometric model to be optimized, so as to obtain the internal thermal fluid flow and heat dissipation state of the radiator.

[0050] According to the incompressible steady-state fluid flow, the fluid-thermal coupling control equation of the geometric model includes the mass, momentum, and energy conservation equations in dimensionless form, and its expression is:

[0051]

[0052] In the formula, Re is the Reynolds number, Pr is the Prandtl number, ▽ is the dimensionless gradient operator, the state variable u is the dimensionless velocity, p is the dimensionless pressure, T is the dimensionless temperature, the material parameter α is the inverse permeability, and β is the temperature difference heat transfer coefficient.

[0053] Let Γ in , Γ out , Γ wall represent the fluid inlet, outlet, and wall no-slip boundary respectively, u in , T in be the velocity and temperature at the given fluid inlet, and the fluid velocity and temperature boundary conditions of the fluid-structure system are:

[0054]

[0055] In the formula, I is the identity matrix, and n is the unit normal vector pointing outwards from the boundary.

[0056] The fluid-thermal coupling control equation describes the flow and heat transfer of the fluid-structure system under the constant internal heat source temperature in the computational domain and the given boundary conditions.

[0057] Under the background of the reconstructed body-fitted grid, map the level set function φ to the characteristic function χ φ according to the explicit fluid-structure boundary (i.e., the zero isocontour), and perform a "1 / 0 binary structure" material mapping on the fluid-structure system. The expression of the obtained mapping relationship is:

[0058]

[0059] Based on the obtained mapping relationship, the material interpolation is performed using the Rational Approximation of Material Penalization (RAMP). The formula corresponding to the material interpolation is as follows:

[0060]

[0061] In the formula, q a and q b are interpolation penalty factors, and the value of this method is 0.01; α max is the maximum inverse permeability, which should be set large enough to ensure that the flow velocity in the solid domain is small enough; β max is the maximum temperature difference heat transfer coefficient. Since there is no intermediate density, the material interpolation is for the convenience of sensitivity analysis and solution.

[0062] Step 3: Considering the multi-objectives of maximizing the structural heat dissipation and minimizing the fluid dissipated energy, an adaptive weighted method is used to establish the objective function. Taking the maximum volume fraction and the maximum fluid dissipated energy as inequality constraint conditions, taking the fluid-thermal coupling control equation as the equality constraint condition, and taking the level set function as the design variable to construct the topology optimization model.

[0063] The objectives of the optimization problem are to maximize the structural heat dissipation and minimize the fluid dissipated energy. The adaptive weighted method is used to determine the objective function. The expression of the objective function is:

[0064] J = -w1Q + w2(α·Φ)

[0065] where Q = ∫ D β(χ φ )(1 - T)dΩ, which is the temperature field objective, that is, to maximize the heat generation in the solid region; is the velocity field objective, that is, to minimize the dissipated energy in the flow field; w1 and w2 are weight coefficients, satisfying w1 + w2 = 1; α is the adaptive numerical matching coefficient to ensure that the order of magnitude of the two objective function values is the same.

[0066] Taking the maximum volume fraction and the maximum fluid dissipated energy as constraint conditions, and taking the level set function as the design variable, the optimization formulation of the constructed topology optimization model is:

[0067]

[0068] where G1 is the volume constraint, V max is the upper limit of the allowable volume fraction of the fluid domain, G2 is the dissipated energy constraint, is the upper limit of the allowable fluid dissipated energy, both of which are set according to requirements.

[0069] Step 4: Solve the topological optimization model until convergence, output the level set function corresponding to the convergence, and the structure corresponding to the obtained level set function is the optimized runner structure.

[0070] Due to the introduction of body-fitted grids, the dimension of the design variables changes continuously with grid reconstruction during the optimization process. Classical optimization algorithms in the field of topological optimization, including the optimization criterion method (OC) and the method of moving asymptotes (MMA), cannot be directly applied to this optimization problem.

[0071] In this embodiment, the augmented Lagrangian method is used as the optimization algorithm, and the Lagrangian function is constructed according to the optimization formulation. The corresponding formula is:

[0072]

[0073] In the formula, u a , p a , T a , λ1, and λ2 are Lagrange multipliers.

[0074] To obtain the optimization sensitivity, the continuous adjoint method is used to perform sensitivity analysis on the topological optimization model. According to the Karush-Kuhn-Tucker (KTT) conditions of the partial differential equation constrained optimization problem, the sensitivity of the Lagrangian function is derived to obtain the adjoint equation:

[0075]

[0076] Among them, u A , p A , T A are the adjoint variables to be solved, that is, the Lagrange multipliers corresponding to the equality constraints.

[0077] Furthermore, the sensitivity formula is obtained:

[0078]

[0079] The reaction diffusion equation (RDE) is used to update the level set function. The reaction diffusion equation is divided into two parts: the reaction term and the diffusion term. The reaction term contains the sensitivity of the objective function with respect to the design variables, enabling automatic opening of holes during the topological evolution process. The diffusion term ensures the regularity of the level set function and avoids the traditional level set re-regularization process.

[0080] The reaction diffusion equation is:

[0081]

[0082] Among them, φ is the level set function, t is the virtual evolution time, K is the normalization ratio coefficient, is the optimization sensitivity, and τ is the diffusion coefficient.

[0083] The optimization sensitivity is obtained by the continuous adjoint method, and the magnitude of the diffusion coefficient can adjust the geometric complexity of the topological configuration.

[0084] In the reaction-diffusion equation, corresponding to the time integration, the first-order Euler difference scheme is used to discretize the reaction-diffusion equation, and we get:

[0085]

[0086] Among them, Δt is the virtual time step, φ n is the level set function before update, and φ n+1 is the level set function after update. The reaction-diffusion equation discretized in time is solved by the finite element method. The updated level set needs to be forced to be within the range of [-1, 1] to ensure the stability of the iterative process.

[0087] Please refer to Figure 2 , and the solution steps of the topology optimization model are as follows:

[0088] 1) Define the material parameters and optimization parameters, and construct a body-fitted grid according to the initial level set function.

[0089] 2) Interpolate the material parameters, and establish a binary mapping relationship between the level set and the material properties through the characteristic function. The fluid-thermal coupling control equation is solved under the background of the body-fitted grid, and then the objective function and constraint values are calculated. For the numerical matching coefficient α of the objective function, in this embodiment, an adaptive update strategy is adopted, and α = Q / Φ is defined, which is updated every 20 steps in the iterative process. This strategy can avoid the difference in order of magnitude between the two objectives and ensure the stability of the optimization process.

[0090] 3) Sensitivity analysis. Solve the adjoint equation to obtain the adjoint variables. For the Lagrange multiplier λ corresponding to the inequality constraint in the Lagrangian function, the augmented Lagrangian method is used for update, and the update scheme is:

[0091]

[0092] Among them, n is the number of optimization iteration steps, i is the inequality constraint serial number, c and a are constant coefficients, which are set to 1.0 in this embodiment, is the intermediate variable of the update strategy. To stabilize the optimization process, in the first 10 steps, the constraint conditions are relaxed, and then changed to accurate constraints until convergence. The constraint relaxation strategy is:

[0093]

[0094] Among them, G imax is the maximum volume fraction or the maximum fluid dissipation energy in the optimized column type, and n0 is taken as 10.

[0095] 3) Solve the reaction-diffusion equation according to the optimization sensitivity, update the level set function, and set the level set function value of the non-design domain to 1.0.

[0096] 4) Reconstruct the body-fitted grid according to the updated level set function, and return to step 2) until the optimization process converges. Finally, the optimal flow channel structure of the radiator that meets the requirements and has a clear and smooth fluid-solid boundary is obtained.

[0097] The following takes a specific embodiment to further elaborate on the present invention in detail.

[0098] The embodiment of the present invention provides a method for topologically optimizing the internal flow channel of a liquid-cooled radiator for electronic devices. The design space of the radiator is square, and the heat source transfers heat to the plane at a constant temperature. A two-dimensional model is used to simplify the calculation. The simplified design model is as Figure 3 shown. The parameters are all in dimensionless form. The length of the design domain is 8 and the width is 8. To ensure the connectivity of the flow channel during the iteration process, rectangular non-design domains are set at the fluid inlet and outlet, which are always fluid domains here. At the inlet boundary, the maximum flow velocity is u = 1, and it is distributed in a parabolic shape along the inlet boundary. The temperature is set to T = 0; at the outlet boundary, the pressure is evenly distributed, and the magnitude is set to p = 0. The temperature condition is set to an adiabatic condition, that is, n·▽T = 0; due to the symmetry of the design space, half of the design space is taken during the optimization process to simplify the calculation, and symmetric boundary conditions are applied at the symmetric boundary; other boundaries are set to avoid the no-slip boundary condition, that is, u = 0. The Reynolds number is set to Re = 20, the Prandtl number is Pr = 6.7, the maximum inverse permeability is α max = 10 4 , and the maximum temperature difference heat transfer coefficient is β max = 10.

[0099] In this example, the maximum fluid domain volume fraction V max = 0.45, the maximum flow field dissipation energy is Φ = 20, the temperature weight w1 in the objective function is 0.9, the flow field weight w1 is 0.1, the maximum number of iteration steps is set to 400 steps, and the iteration convergence condition is that the relative error of the objective function is less than 10 -4 . In addition, the parameter settings related to the body-fitted grid include the maximum grid size h max = 0.1, the minimum grid size h min = 0.001, the grid change gradient hgrad = 1.2, and the Hausdorff distance hausd = 0.05.

[0100] The iterative curve obtained through the optimization iteration by the method of the present invention is as follows Figure 4 shown. The blue curve represents the objective function, the red curve represents the fluid dissipation energy constraint, and the black curve represents the volume fraction constraint. It can be seen that the entire iterative process tends to be stable after 100 steps, and finally the optimization result as shown in Figure 5 is obtained, which has an explicit, clear and smooth fluid-solid boundary, and can be directly processed and manufactured without cumbersome post-processing procedures.

[0101] In summary, in the optimization process of the present invention, the body-fitted grid technology is adopted to adaptively reconstruct the model grid, obtaining a clear and smooth fluid-solid boundary. At the same time, based on the body-fitted grid, a binary structure material mapping method is used to model the fluid-solid system, improving the solution accuracy. In the problem of optimizing the radiator flow channel design, two objectives of heat dissipation and flow resistance are comprehensively considered, and the weights can be autonomously assigned to guide the radiator design in specific scenarios. In addition, the continuous adjoint method is used for sensitivity analysis to ensure the accuracy of the optimization sensitivity, and the level set function is updated by solving the reaction-diffusion equation to realize automatic opening of holes, ensuring the regularity of the level set function, thereby enhancing the stability of the optimization iteration.

[0102] The present invention also provides a fluid-thermal coupling topology optimization design system for a radiator flow channel based on a body-fitted grid. The system includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it executes the above-mentioned method for optimizing the structure topology of a radiator flow channel based on a body-fitted grid.

[0103] The present invention also provides a computer-readable storage medium. The computer-readable storage medium stores machine-executable instructions. When the machine-executable instructions are called and executed by a processor, the machine-executable instructions cause the processor to implement the above-mentioned method for optimizing the structure topology of a radiator flow channel based on a body-fitted grid.

[0104] It is easy for those skilled in the art to understand that the above are only the preferred embodiments of the present invention, and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A topology optimization design method for a heat sink flow channel structure based on a body-fitted grid, characterized in that: The method comprises the following steps: (1) Based on the initial level set function of the heat sink to be optimized, a body-fitting grid that fits the fluid-solid interface is constructed according to the zero contour of the level set function; (2) Perform binary material mapping of the fluid-solid system according to the fluid-solid boundary, and use the obtained mapping relationship to solve the fluid-heat coupling control equation of the geometric model to be optimized to obtain the thermal fluid flow and heat dissipation state inside the radiator; (3) Considering the multi-objectives of maximizing structural heat dissipation and minimizing fluid dissipated energy, an adaptive weighted method is used to establish the objective function, with the maximum volume fraction and the maximum fluid dissipated energy as inequality constraints, the fluid-heat coupling control equation as the equality constraint, and the level set function as the design variable, to construct a topology optimization model based on body-fitting mesh; (4) Solving the topology optimization model until convergence, outputting the corresponding level set function at the time of convergence, and the structure corresponding to the obtained level set function is the optimized flow channel structure.

2. The method for topological optimization design of a heat sink flow channel structure based on a body-fitted grid according to claim 1, characterized in that: The level set function φ describes the fluid phase and solid phase of the geometric model respectively. By updating the level set function value, the material distribution of the fluid-solid system is changed, thereby obtaining the flow channel structure expressed by the level set function. The mapping relationship between the level set function and the material distribution of the fluid-solid system in the geometric model is: In the formula, D represents the model calculation domain, Ω represents the fluid domain, and the complementary domain D\Ω represents the solid domain. represents the fluid-solid boundary.

3. The method for topological optimization design of a heat sink flow channel structure based on body-fitted mesh according to claim 1, characterized in that: The body-fitting grid technology is used to adaptively reconstruct the grid of the geometric model. The grid of the geometric model is divided into different sub-grids with explicit boundaries according to the zero contour of the level set function. The divided sub-grids represent the solid phase and fluid phase of the geometric model, respectively, realizing the explicit expression of the fluid-solid interface at the grid level.

4. The method for topological optimization design of a heat sink flow channel structure based on body-fitted mesh according to claim 1, characterized in that: The fluid-heat coupling control equation is: Where Re is the Reynolds number, Pr is the Prandtl number, ▽ is the dimensionless gradient operator, the state variable u is the dimensionless velocity, p is the dimensionless pressure, T is the dimensionless temperature, the material parameter α is the reverse permeability, and β is the temperature difference heat transfer coefficient.

5. The method for topological optimization design of a heat sink flow channel structure based on body-fitted mesh according to claim 1, characterized in that: In the context of the reconstructed body-fitting grid, the level set function φ is mapped to the characteristic function χ according to the explicit fluid-solid boundary. φ On the surface, the material mapping of the 1 / 0 binary structure is performed on the fluid-solid system, and the expression of the mapping relationship is obtained as follows: Based on the obtained mapping relationship, the material property rational approximation model is used for material interpolation. The formula corresponding to the material interpolation is: In the formula, q a ,q b is the interpolation penalty factor; α max is the maximum reverse osmosis rate; β max is the maximum temperature difference heat transfer coefficient.

6. A method for topological optimization design of fluid-heat multi-physics field coupling based on body-fitted grid as described in claim 1, characterized in that: The reaction-diffusion equation is used to update the level set function. The reaction-diffusion equation is divided into two parts: the reaction term and the diffusion term. The reaction term contains the sensitivity of the objective function to the design variables, so that holes are automatically opened during the topological evolution process. The diffusion term ensures the regularity of the level set function. The reaction-diffusion equation is: Among them, φ is the level set function, t is the virtual evolution time, K is the normalized scale coefficient, To optimize sensitivity, τ is the diffusion coefficient and n is the unit normal vector outward from the boundary.

7. The method for topological optimization design of a heat sink flow channel structure based on a body-fitted grid according to any one of claims 1 to 6, characterized in that: The optimization formula of the topology optimization model is: Findφ Max J=-w1Q+w2(α·Φ) Among them, G1 is the volume constraint, V max is the upper limit of the allowed fluid domain volume fraction, G2 is the dissipated energy constraint, is the upper limit of the energy dissipated by the fluid.

Images