Topological optimization method of variable height reinforcement based on Sigmoid function

Through the topology optimization method of variable height reinforcement based on Sigmoid function, the reinforcing shape is optimized, which solves the limitation problem that only equal height reinforcement can be obtained in the existing technology, and achieves good stiffness transmission of variable height reinforcement structures Performance and process achievability.

CN114282418BActive Publication Date: 2024-11-05AECC HUNAN AVIATION POWERPLANT RES INST +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202111617724.7
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2021-12-27
Publication Date
2024-11-05
Estimated Expiration
2041-12-27

AI Technical Summary

Technical Problem

The existing topological optimization methods generally only optimize the distribution form of reinforcement, and lack optimization methods for reinforcement shapes, resulting in only equal height reinforcement optimization results, which have great limitations.

Method used

The final unit density vector is obtained by using the variable height reinforcement topology optimization method based on Sigmoid function. The reinforcement height is achieved by establishing a finite element model, constructing the Helmholtz anisotropic filtering function, and performing linear and nonlinear filtering. optimization.

Benefits of technology

The effect of variable height reinforcement is achieved, and the structural form with good stiffness transfer performance is obtained, which is easy to operate and meets process realization.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114282418B_ABST
    Figure CN114282418B_ABST
Patent Text Reader

Abstract

The present invention discloses a variable height reinforcement topology optimization method based on Sigmoid function, which includes establishing a finite element model of a variable height rib structure design domain, dividing it into finite element grids; setting a unit set according to the finite element grid boundary, load conditions, and rib forming direction; constructing a Helmholtz anisotropic filter function based on parameters in the unit set; solving the weak form of the Helmholtz anisotropic filter function by a finite element method; solving the filtered unit density vector by the weak form of the Helmholtz anisotropic filter function, that is, performing linear filtering; performing nonlinear filtering on the density after linear filtering based on an inverse S-type function to obtain a final filtered density; and establishing a topology optimization result based on the final filtered density. The present invention filters the rib height by a nonlinear mapping method, can achieve the purpose of freely controlling the reinforcement height, and can finally obtain a variable height reinforcement structure with good stiffness transfer performance.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of engineering structure optimization design, and in particular to a variable height reinforcement topology optimization method based on a Sigmoid function. Background Art

[0002] This patent uses a topological optimization method to perform conceptual design on thin-walled reinforced structures. When conducting innovative structural design, it is necessary to combine process feasibility, so the topological optimization method needs to consider the impact of manufacturing process constraints. In order to meet the requirements of aerodynamic shape, confined space, strength, stiffness, etc., thin-walled structures in engineering are often designed as thin-walled reinforced structures. The shape of the reinforcement (reinforcement height, reinforcement cross-sectional shape, etc.) and its distribution on the skin directly determine the mechanical properties of the structure. Thin-walled reinforced structures are generally processed and manufactured by mechanical milling, extrusion, and more advanced additive manufacturing processes.

[0003] The mechanical properties of thin-walled reinforced structures are closely related to the shape and distribution of the reinforcement. Existing topological optimization methods generally only optimize the distribution of the reinforcement, but lack methods for optimizing the shape of the reinforcement. Summary of the invention

[0004] In view of the above-mentioned deficiencies in the prior art, the variable height reinforcement topology optimization method based on Sigmoid function provided by the present invention solves the problems that the prior art cannot freely control the reinforcement height and cannot achieve good stiffness transferability.

[0005] In order to achieve the above-mentioned object of the invention, the technical solution adopted by the present invention is:

[0006] A variable height reinforcement topology optimization method based on a Sigmoid function is provided, which comprises the following steps:

[0007] S1. Establish a finite element model of the design domain of the variable height rib structure and divide it into finite element grids;

[0008] S2, setting a unit set according to the finite element mesh boundary, load conditions, and rib forming direction;

[0009] S3, constructing a Helmholtz anisotropic filtering function based on the parameters in the unit set;

[0010] S4, solving the Helmholtz anisotropic filter function using the finite element method to obtain the weak form of the Helmholtz anisotropic filter function;

[0011] S5, solving the filtered unit density vector by using the weak form of the Helmholtz anisotropic filtering function, that is, performing linear filtering;

[0012] S6, performing nonlinear filtering on the linearly filtered unit density vector based on an inverse S-type function to obtain a final filtered unit density vector;

[0013] S7. Based on the final filter unit density vector, a topology optimization result is established.

[0014] Furthermore, the specific process in step S2 is:

[0015] Create a unit set according to the finite element mesh boundary, load conditions and rib forming direction, the unit set includes: design domain unit set, non-design domain unit set, load node set, boundary node set, rib root node set, rib top node set;

[0016] According to the design domain unit set, non-design domain unit set, load node set and boundary node set in the unit set, the stiffness matrix is ​​assembled and the finite element solution is performed to obtain the overall stiffness matrix;

[0017] The height growth direction of the rib is determined according to the rib root node set and the rib top node set in the unit set: the rib root node set is assigned to 0, and the rib top node set is assigned to 1, a variable field is constructed, and the diffusion equation is used to solve the direction vector of the variable field, that is, the extrusion direction of the Helmholtz anisotropic filtering function.

[0018] Furthermore, the specific process of constructing the Helmholtz anisotropic filtering function in step S3 is:

[0019] According to the formula:

[0020]

[0021]

[0022] Construct a Helmholtz anisotropic filter function where is the density field after filtering by anisotropic filtering function, ρ is the density field before filtering, n is the normal vector of the design domain boundary Γ, Ω is the design domain, is a vector differential operator, on is inclusive, and c is a 3×3 second-order positive definite tensor used to describe the diffusion effect and to determine the range of filtering:

[0023]

[0024] V=[v n ,v t1 ,v t2 ]

[0025] Where V is the matrix form of the spatial basis vector in the three directions of the local coordinate system, T is the transpose, r n v is the reinforcement height direction nThe filter radius, r t1 is the extrusion direction v t1 The filter radius, r t2 is the extrusion direction v t2 The filter radius, v n 、v t1 and r t2 Mutually orthogonal.

[0026] Furthermore, the specific process of step S4 is:

[0027] According to the formula:

[0028]

[0029]

[0030] Convert the Helmholtz anisotropic filter function to a weak form; where δ is the symbol of the variation.

[0031] Furthermore, the specific process of step S5 is:

[0032] S5-1. Based on the weak form of the Helmholtz anisotropic filtering function, the transformation between node variables and unit variables is performed according to the formula:

[0033]

[0034] ρ q =∫N T ρdΩ q

[0035] The unit coefficient matrix H of the Helmholtz equation after finite element discretization is obtained q and the right end vector ρ of the unit node q ; where N is the unit shape function, Ω q is the set of units in the filter area centered on unit q;

[0036] S5-2, according to the formula:

[0037]

[0038] Get the filtered cell density vector That is, the linear output of the weak form of the Helmholtz equation after filtering the ribs; where H is the linear output of H q The total coefficient matrix, T and T * are all transformation matrices, T is composed of ∫N T dΩ q Composition, T * By ∫NdΩ q / ∫dΩ q constitute.

[0039] Furthermore, the specific process of obtaining the final filtration density in step S6 is as follows:

[0040] According to the formula:

[0041]

[0042]

[0043] α n+1 =α n -0.1

[0044] Get the final filter unit density vector Where e is the logarithm of natural numbers, s is the normalized parameter coordinate in the draft direction, β is the n+1th penalty factor, log is the logarithmic function, ψ is the parameter used to adjust the degree of nonlinear filtering, and α n+1 is the n+1th iteration factor, α n is the nth iteration factor, and n is the number of iterations.

[0045] Furthermore, the topology optimization result in step S7 is:

[0046]

[0047]

[0048]

[0049] 0<ρ min ≤ρ q ≤1,q=1,...,N

[0050] Where f is the response function of topology optimization, K is the overall stiffness matrix, and the final filtered element density vector Calculated, F is the external load vector, U is the displacement vector, V f is the volume ratio constraint, ρ min is the minimum density, where the subscript q represents the corresponding physical quantity of the qth unit, is the physical density of unit q, ρ q is the pseudo-density of unit q, v q is the volume of unit q, and N is the number of units.

[0051] The beneficial effects of the present invention are:

[0052] The present invention aims at the problem that the existing topology optimization methods generally only optimize the distribution form of reinforcement, but lack the optimization method for the reinforcement shape, which leads to the optimization result that only the equal-height reinforcement can be obtained, and the optimization problem has great limitations. The present invention provides a variable-height reinforcement topology optimization method based on anisotropic filtering. The height of the reinforcement is filtered by the method of the present invention, so that the variable-height reinforcement effect can be achieved, and finally a variable-height reinforcement structure with good stiffness transfer performance can be obtained. The method proposed by the present invention is simple to operate, meets the process feasibility, and is easy to be integrated into the topology optimization program. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 is a flow chart of the present invention;

[0054] Figure 2 It is an inverse S-shaped function curve;

[0055] Figure 3 This is a schematic diagram of the application of variable height reinforced structure engineering;

[0056] Figure 4 This is a schematic diagram of the application of uniform height reinforced structure engineering;

[0057] Figure 5 This is an oblique view of the variable height reinforcement example diagram;

[0058] Figure 6 This is a schematic side view of the variable height reinforcement example. DETAILED DESCRIPTION

[0059] The specific implementation modes of the present invention are described below so that those skilled in the art can understand the present invention. However, it should be clear that the present invention is not limited to the scope of the specific implementation modes. For those of ordinary skill in the art, as long as various changes are within the spirit and scope of the present invention as defined and determined by the attached claims, these changes are obvious, and all inventions and creations utilizing the concept of the present invention are protected.

[0060] like Figure 1 and Figure 2 As shown, the variable height reinforcement topology optimization method based on Sigmoid function includes the following steps:

[0061] S1. Establish a finite element model of the design domain of the variable height rib structure and divide it into finite element grids;

[0062] S2, setting a unit set according to the finite element mesh boundary, load conditions, and rib forming direction;

[0063] S3, constructing a Helmholtz anisotropic filtering function based on the parameters in the unit set;

[0064] S4, solving the Helmholtz anisotropic filter function using the finite element method to obtain the weak form of the Helmholtz anisotropic filter function;

[0065] S5, solving the filtered unit density vector by using the weak form of the Helmholtz anisotropic filtering function, that is, performing linear filtering;

[0066] S6, performing nonlinear filtering on the linearly filtered unit density vector based on an inverse S-type function to obtain a final filtered unit density vector;

[0067] S7. Based on the final filter unit density vector, a topology optimization result is established.

[0068] The specific process in step S2 is:

[0069] Create a unit set according to the finite element mesh boundary, load conditions and rib forming direction, the unit set includes: design domain unit set, non-design domain unit set, load node set, boundary node set, rib root node set, rib top node set;

[0070] According to the design domain unit set, non-design domain unit set, load node set and boundary node set in the unit set, the stiffness matrix is ​​assembled and the finite element solution is performed to obtain the overall stiffness matrix;

[0071] The height growth direction of the rib is determined according to the rib root node set and the rib top node set in the unit set: the rib root node set is assigned to 0, and the rib top node set is assigned to 1, a variable field is constructed, and the diffusion equation is used to solve the direction vector of the variable field, that is, the extrusion direction of the Helmholtz anisotropic filtering function.

[0072] The specific process of constructing the Helmholtz anisotropic filtering function in step S3 is:

[0073] According to the formula:

[0074]

[0075]

[0076] Construct a Helmholtz anisotropic filter function where is the density field after filtering by anisotropic filtering function, ρ is the density field before filtering, n is the normal vector of the design domain boundary Γ, Ω is the design domain, is a vector differential operator, on is inclusive, and c is a 3×3 second-order positive definite tensor used to describe the diffusion effect and to determine the range of filtering:

[0077]

[0078] V=[vn ,v t1 ,v t2 ]

[0079] Where V is the matrix form of the spatial basis vector in the three directions of the local coordinate system, T is the transpose, r n v is the reinforcement height direction n The filter radius, r t1 is the extrusion direction v t1 The filter radius, r t2 is the extrusion direction v t2 The filter radius, v n 、v t1 and r t2 Mutually orthogonal.

[0080] The specific process of step S4 is:

[0081] According to the formula:

[0082]

[0083]

[0084] Convert the Helmholtz anisotropic filter function to a weak form; where δ is the symbol of the variation.

[0085] The specific process of step S5 is:

[0086] S5-1. Based on the weak form of the Helmholtz anisotropic filtering function, the transformation between node variables and unit variables is performed according to the formula:

[0087]

[0088] ρ q =∫N T ρdΩ q

[0089] The unit coefficient matrix H of the Helmholtz equation after finite element discretization is obtained q and the right end vector ρ of the unit node q ; where N is the unit shape function, Ω q is the set of units in the filter area centered on unit q;

[0090] S5-2, according to the formula:

[0091]

[0092] Get the filtered cell density vector That is, the linear output of the weak form of the Helmholtz equation after filtering the ribs; where H is the linear output of H q The total coefficient matrix, T and T* are all transformation matrices, T is composed of ∫N T dΩ q Composition, T * By ∫NdΩ q / ∫dΩ q constitute.

[0093] The specific process of obtaining the final filtration density in step S6 is:

[0094] According to the formula:

[0095]

[0096]

[0097] α n+1 =α n -0.1

[0098] Get the final filter unit density vector Where e is the logarithm of natural numbers, s is the normalized parameter coordinate in the draft direction, β is the n+1th penalty factor, log is the logarithmic function, ψ is the parameter used to adjust the degree of nonlinear filtering, and α n+1 is the n+1th iteration factor, α n is the nth iteration factor, and n is the number of iterations.

[0099] The topology optimization result in step S7, i.e., the topology optimization formula under the minimum compliance objective is

[0100] for:

[0101]

[0102]

[0103]

[0104] 0<ρ min ≤ρ q ≤1,q=1,...,N

[0105] Where f is the response function of topology optimization, K is the overall stiffness matrix, and the final filtered element density vector Calculated, F is the external load vector, U is the displacement vector, V f is the volume ratio constraint, ρ min is the minimum density, where the subscript q represents the corresponding physical quantity of the qth unit, is the physical density of unit q, ρ q is the pseudo-density of unit q, v q is the volume of unit q, and N is the number of units.

[0106] like Figure 3 and Figure 4 As shown, the ribs of the present invention are of variable height, which is different from the ribs of constant height in the prior art.

[0107] like Figure 5 and Figure 6 As shown, the present invention can realize the variable height reinforcement design on the special-shaped curved surface.

[0108] The present invention aims at the problem that the existing topology optimization methods generally only optimize the distribution form of reinforcement, but lack the optimization method for the reinforcement shape, resulting in only the optimization result of equal-height reinforcement, and the optimization problem has great limitations. The present invention provides a variable-height reinforcement topology optimization method based on anisotropic filtering. The height (density distribution) of the reinforcement is filtered by the method of the present invention, so that the variable-height reinforcement effect can be achieved, and finally a variable-height reinforcement structure with good stiffness transfer performance can be obtained. The method proposed by the present invention is simple to operate, meets the process feasibility, and is easy to integrate into the topology optimization program.

Claims

1. A variable height reinforcement topology optimization method based on Sigmoid function, It is characterized in that The following steps are involved: S1. Establish a finite element model of the design domain of the variable height rib structure and divide it into finite element grids; S2, setting a unit set according to the finite element mesh boundary, load conditions, and rib forming direction; S3, constructing a Helmholtz anisotropic filtering function based on the parameters in the unit set; S4, solving the Helmholtz anisotropic filter function using the finite element method to obtain the weak form of the Helmholtz anisotropic filter function; S5, solving the filtered unit density vector by using the weak form of the Helmholtz anisotropic filtering function, that is, performing linear filtering; S6, performing nonlinear filtering on the linearly filtered unit density vector based on an inverse S-type function to obtain a final filtered unit density vector; S7. Based on the final filter unit density vector, a topology optimization result is established.

2. According to the variable height reinforcement topology optimization method based on Sigmoid function in claim 1, It is characterized in that The specific process in step S2 is: Create a unit set according to the finite element mesh boundary, load conditions and rib forming direction, the unit set includes: design domain unit set, non-design domain unit set, load node set, boundary node set, rib root node set, rib top node set; According to the design domain unit set, non-design domain unit set, load node set and boundary node set in the unit set, the stiffness matrix is ​​assembled and the finite element solution is performed to obtain the overall stiffness matrix; The height growth direction of the rib is determined according to the rib root node set and the rib top node set in the unit set: the rib root node set is assigned to 0, and the rib top node set is assigned to 1, a variable field is constructed, and the diffusion equation is used to solve the direction vector of the variable field, that is, the extrusion direction of the Helmholtz anisotropic filtering function.

3. According to claim 2, the variable height reinforcement topology optimization method based on Sigmoid function, It is characterized in that The specific process of constructing the Helmholtz anisotropic filtering function in step S3 is: According to the formula: Construct a Helmholtz anisotropic filter function where is the density field after filtering by anisotropic filtering function, ρ is the density field before filtering, n is the normal vector of the design domain boundary Γ, Ω is the design domain, is a vector differential operator, on is inclusive, and c is a 3×3 second-order positive definite tensor used to describe the diffusion effect and to determine the range of filtering: V=[v n ,v t1 ,v t2 ] Where V is the matrix form of the spatial basis vector in the three directions of the local coordinate system, T is the transpose, r n v is the reinforcement height direction n The filter radius, r t1 is the extrusion direction v t1 The filter radius, r t2 is the extrusion direction v t2 The filter radius, v n 、v t1 and r t2 Mutually orthogonal.

4. According to claim 3, the variable height reinforcement topology optimization method based on Sigmoid function, It is characterized in that The specific process of step S4 is: According to the formula: Convert the Helmholtz anisotropic filter function to a weak form; where δ is the symbol of the variation.

5. According to claim 4, the variable height reinforcement topology optimization method based on Sigmoid function, It is characterized in that The specific process of step S5 is: S5-1. Based on the weak form of the Helmholtz anisotropic filtering function, the transformation between node variables and unit variables is performed according to the formula: r q =∫N T ρdΩ q The unit coefficient matrix H of the Helmholtz equation after finite element discretization is obtained q and the right end vector ρ of the unit node q ; where N is the unit shape function, Ω q is the set of units in the filter area centered on unit q; S5-2, according to the formula: Get the filtered cell density vector That is, the linear output of the weak form of the Helmholtz equation after filtering the ribs; where H is the linear output of H q The total coefficient matrix, T and T * are all transformation matrices, T is composed of ∫N T dΩ q Composition, T * By ∫NdΩ q / ∫dΩ q constitute.

6. The variable height reinforcement topology optimization method based on Sigmoid function according to claim 5, It is characterized in that The specific process of obtaining the final filtration density in step S6 is: According to the formula: α n+1 =α n -0.1 Get the final filter unit density vector Where e is the logarithm of natural numbers, s is the normalized parameter coordinate in the draft direction, β is the n+1th penalty factor, log is the logarithmic function, ψ is the parameter used to adjust the degree of nonlinear filtering, and α n+1 is the n+1th iteration factor, α n is the nth iteration factor, and n is the number of iterations.

7. The variable height reinforcement topology optimization method based on Sigmoid function according to claim 6, It is characterized in that The topology optimization result in step S7 is: 0<ρ min ≤ρ q ≤1,q=1,...,N Where f is the response function of topology optimization, K is the overall stiffness matrix, and the final filtered element density vector Calculated, F is the external load vector, U is the displacement vector, V f is the volume ratio constraint, ρ min is the minimum density, where the subscript q represents the corresponding physical quantity of the qth unit, is the physical density of unit q, ρ q is the pseudo-density of unit q, v q is the volume of unit q, and N is the number of units.