A bidirectional evolutionary structural optimization algorithm for aerospace structures
By using a two-way progressive structural optimization algorithm to determine the evolution rate in aerospace structures, the problems of increasing the number of iteration steps and not converging the structure in traditional methods are solved, and the topological optimization of the load design correlation is realized, which improves the optimization efficiency and accuracy.
Patent Information
- Application Number
- CN202410057950.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-01-15
- Publication Date
- 2024-11-15
- Estimated Expiration
- 2044-01-15
AI Technical Summary
The traditional two-way progressive structural optimization method has problems such as increasing the number of iteration steps and not converging the structure in aerospace structures, especially when considering non-design-related loads, the errors and structural unreasonable problems caused by design-related loads are not effectively optimized.
A bidirectional progressive structure optimization algorithm is used to determine the evolution rate through the change function, consider the design correlation of loads, reduce the number of iteration steps, and improve the optimization efficiency. The specific steps include selecting the structure, establishing a finite element model, applying loads, calculating sensitivity, and updating Structural volume and material parameters until convergence criteria are met.
It realizes the reduction of optimization iteration steps while ensuring calculation accuracy, improve optimization efficiency, ensures the accuracy and reliability of topological optimization results under design-related loads, and avoids the problems of non-convergence of structure and low efficiency.
Smart Images

Figure CN117875127B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of aerospace technology, and in particular to a bidirectional progressive structural optimization algorithm that can be used for aerospace structures. Background Art
[0002] Topology optimization is a structural optimization method that seeks the optimal distribution of materials within a design region under given loads, constraints, and performance indicators. It is widely used in aerospace and other fields. The bidirectional progressive structural optimization method is one of the most widely used topology optimization design methods. However, the traditional bidirectional progressive structural optimization method uses a fixed evolution rate. If the selected evolution rate is small, the number of iterations will increase and the optimization efficiency will decrease. If a larger evolution rate is used, the structure may not converge.
[0003] The loads in the topology optimization design domain can be divided into design-related loads and non-design-related loads. For design-related loads, their characteristics are coupling with the structure. As the structure changes, the load will also change, such as deadweight, centrifugal load, etc. For non-design-related loads, their characteristics are that they do not change with the structure and are always fixed during the optimization process. Aircraft engine parts such as turbine disks are mainly affected by centrifugal loads. Failure to consider design-related loads will lead to large errors and unreasonable structures.
[0004] Therefore, it is necessary to provide a bidirectional progressive structural optimization algorithm that can be used for aerospace structures, which can reduce the number of iterations required for optimization while ensuring the calculation accuracy and improve the optimization efficiency. Summary of the invention
[0005] The present invention discloses a bidirectional progressive structural optimization algorithm that can be used for aerospace structures. The method takes into account the design correlation of loads, uses a variation function to determine the evolution rate, and reduces the number of iteration steps required for topology while ensuring good convergence performance, thereby improving the optimization efficiency. The method can effectively solve the technical problems involved in the background technology.
[0006] To achieve the above object, the technical solution of the present invention is:
[0007] A bidirectional progressive structural optimization algorithm for aerospace structures includes the following steps:
[0008] S1. Select the structure to be optimized and determine the load type;
[0009] S2. Establish a finite element model and perform meshing;
[0010] S3, assign material parameters, and apply boundary conditions and loads;
[0011] S4, determining optimization parameters of the bidirectional progressive structural optimization method;
[0012] S5. Perform finite element analysis and calculate sensitivity;
[0013] S6, calculating the evolution rate of the current iteration step and determining the updated structure volume;
[0014] S7, update the unit and structure, and reassign material parameters;
[0015] S8: Determine whether the convergence criteria and constraints are met. If not, return to step S5; if satisfied, end the optimization and output the model.
[0016] A structural optimization algorithm comprises the following specific steps: 1) selecting an optimized structure and analyzing the load type; 2) simplifying the model, establishing a finite element model of the structure (the structure can be divided using a plane four-node rectangular grid), and dividing the design area and the non-design area; 3) assigning material properties, applying boundary conditions and loads; 4) determining the optimization parameters of a related bidirectional progressive structural optimization method such as a penalty factor and a filter radius; 5) performing finite element analysis on the structure, and selecting a sensitivity calculation formula to obtain unit sensitivity according to the applied load type; 6) determining the evolution rate of the current iteration process by a variation function, calculating the updated structure volume in the iteration step, and updating the unit; 7) using a material interpolation model to reassign material properties to the updated unit; 8) updating the finite element model, and judging whether the constraint conditions and the convergence conditions are met, if they are met, terminating the optimization, otherwise returning to step 5); the invention takes into account the design-related characteristics of the load, realizes the topological optimization of the structure under the external load and the design-related load, adopts a variation function to determine the evolution rate, reduces the number of iteration steps required for the optimization while ensuring the calculation accuracy, and improves the optimization efficiency.
[0017] As a preferred improvement of the present invention: the load types include design-related loads and external loads, and the design-related loads include gravity loads and centrifugal loads; when determining the load type, ignore the loads with large magnitude differences and small impact on the results. Specifically, select the structure to be optimized, analyze its main load types through experiments or data query, and then select the main loads, including design-related loads and external loads, and ignore the loads with large magnitude differences and small impact on the results. It can be determined according to the actual situation. Generally speaking, if the load magnitude differs by 10 3 Can be ignored.
[0018] As a preferred improvement of the present invention: in the step S2, the design area and the non-design area are also selected. The non-design area generally refers to the part that is not optimized but must participate in the overall analysis, usually some connection points, or hole parts with specific functions.
[0019] As a preferred improvement of the present invention: in the step S3, the material parameters include Young's modulus, density and Poisson's ratio; in the step S4, the optimization parameters include filter radius, penalty factor, maximum number of iteration steps, volume constraint and convergence judgment. The filter radius is determined based on experience and is generally greater than 1.5. The penalty factor is selected in the range of 5-120, which is also determined based on experience. Data show that 5-8 has a better effect. The maximum number of iteration steps is determined based on actual conditions. The number of steps can be more for complex structures and less for simple structures. The volume constraint refers to the limit on the total volume after optimization. It can be selected according to actual needs. If the total volume of the original structure is 1 and you want to achieve a 30% weight reduction, set it to 0.7. Generally speaking, it is between 0-1. The convergence judgment is generally selected to be 0.001.
[0020] As a preferred improvement of the present invention: in the step S5, the sensitivity is calculated under the objective function and the constraint condition, the objective function is the minimum average flexibility of the structure, and the constraint condition is the volume constraint;
[0021] The established optimization model is as follows:
[0022]
[0023] Where, the subscript i represents the unit number, i = 1, 2, 3...N, X is the set of design variables, x i is the structural design variable of the i-th unit, which takes the value of 1 or the set minimum value x min , C(X) is the objective function, i.e. the flexibility of the structure, F T is the transpose of the force matrix, U is the displacement matrix, V * is the target volume, v i is the volume of the i-th unit;
[0024] If the applied load is an external load, the sensitivity calculation formula is:
[0025]
[0026] Wherein, the subscript i represents the unit number, i = 1, 2, 3...N, α i is the sensitivity of the i-th unit, q is the penalty factor, C is the average flexibility of the structure, x i is the structural design variable of the ith unit, is the structural stiffness matrix of the ith unit, u i is the structural displacement vector of the ith unit, is the transposed matrix of the structural displacement vector of the i-th unit;
[0027] When the applied load is a load related to the design structure, such as gravity load and centrifugal load, the characteristics of this type of load are that the size, direction, and presence or absence of the load are related to the structure;
[0028] If the applied load is a centrifugal load, the sensitivity calculation formula is:
[0029]
[0030] Where ω is the applied speed, r i is the rotation radius of the ith unit, v i is the volume of the i-th unit, ρ0 is the initial density, is the direction matrix of the force;
[0031] If the applied load is a gravity load, the sensitivity calculation formula is:
[0032]
[0033] Where g is the acceleration due to gravity.
[0034] As a preferred improvement of the present invention: in the step S6, the evolution rate and the updated structure volume are calculated by the change function;
[0035] The evolution rate of the kth iteration is:
[0036]
[0037] Then the volume after the kth iteration is
[0038] Among them, SR k is the evolution rate of the current k-th iteration step, SR max is the maximum evolution rate set, SR min is the minimum evolution rate set, S(y) is the change function, s1 and s2 are the control parameters set, y is the ratio of the current volume to the original volume, v is the volume constraint set, V k is the volume after the kth iteration, V k-1 is the volume after the k-1th iteration, V * is the target volume. This function can limit the evolution rate of each superposition to between the set maximum evolution rate and minimum evolution rate, and change with the current volume fraction.
[0039] As a preferred improvement of the present invention: in the step S7, the unit and structure are updated by sensitivity, and the material parameters are reassigned by using the material interpolation model. The material interpolation model is as follows:
[0040]
[0041] Wherein, the subscript i represents the unit number, i = 1, 2, 3...N, ρ(x i) is the density of the ith unit, x i is the structural design variable of the i-th unit, ρ0 is the initial density, E(x i ) is the Young’s modulus of the ith unit, E0 is the initial Young’s modulus, q is the penalty factor, and the interpolation model of the material is the rational approximation model of the material.
[0042] As a preferred improvement of the present invention: in the step S8, it is determined whether the objective function satisfies the convergence standard and the constraint condition, the objective function is that the average flexibility of the structure is minimized, that is, the stiffness of the structure is maximized, the constraint condition is the volume constraint, and the convergence standard is that the average flexibility change of the structure five times before and after is not greater than the set value, as shown in the following formula:
[0043]
[0044] Where C is the average flexibility of the structure, the subscript represents the number of iterations, k represents the kth iteration, j is the count, i.e., five times before and after, and e is the set value of the convergence standard, i.e., the absolute value of the difference between the average flexibility of the structure from the 9th to the 5th iteration before this iteration and the average flexibility of the structure from the 4th to this iteration before this iteration, divided by the average flexibility of the structure from the 4th to this iteration before this iteration, and the result is less than or equal to the set value.
[0045] A bidirectional progressive structural optimization algorithm that can be used for aerospace structures can be applied to the topology optimization of structures subjected to centrifugal loads, wherein in step S1, a turbine disk structure is taken as an example, and its load is mainly a centrifugal load. The finite element model in step S2 refers to the initial structure of the cross section of the turbine disk subjected to centrifugal load that needs to be topologically optimized. After executing steps S1 to S8, an optimized structure can be obtained.
[0046] The beneficial effects of the present invention are as follows:
[0047] The present invention fully considers the design relevance of the load, so that the topology optimization results under the design-related load are more accurate and reliable. A variable evolution rate is achieved by changing the function, which improves the problems of low efficiency caused by a small evolution rate and structural non-convergence caused by a large evolution rate. The optimization efficiency is improved while ensuring good convergence performance. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative work, among which:
[0049] Figure 1A flow chart of a bidirectional progressive structural optimization algorithm that can be used for aerospace structures according to the present invention;
[0050] Figure 2 The finite element model diagram corresponding to the initial cross-sectional structure of the turbine disk of the present invention;
[0051] Figure 3 The structure is a topologically optimized structure of the turbine disk cross section according to the present invention;
[0052] Figure 4 A comparison diagram of the topology optimization process of determining the evolution rate by a variable function and a fixed evolution rate in the present invention. DETAILED DESCRIPTION
[0053] The technical solutions in the embodiments of the present invention will be described clearly and completely below in combination with the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0054] It should be noted that all directional indications in the embodiments of the present invention (such as up, down, left, right, front, back, etc.) are only used to explain the relative position relationship, movement status, etc. between the components under a certain specific posture (as shown in the accompanying drawings). If the specific posture changes, the directional indication will also change accordingly.
[0055] In addition, in the present invention, descriptions such as "first", "second", etc. are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features defined as "first" or "second" may explicitly or implicitly include at least one of the features. In the description of the present invention, the meaning of "plurality" is at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
[0056] In the present invention, unless otherwise clearly specified and limited, the terms "connection", "fixation", etc. should be understood in a broad sense. For example, "fixation" can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.
[0057] In addition, the technical solutions between the various embodiments of the present invention can be combined with each other, but it must be based on the fact that ordinary technicians in the field can implement it. When the combination of technical solutions is contradictory or cannot be implemented, it should be deemed that such combination of technical solutions does not exist and is not within the scope of protection required by the present invention.
[0058] See also Figure 1 As shown, the present invention provides a bidirectional progressive structural optimization algorithm that can be used for aerospace structures, comprising the following steps:
[0059] The first step is to select the structure to be optimized and analyze the main load types of the structure. The main load types are divided into design-related loads and design-irrelevant external loads. Loads with large differences in magnitude and small impact on the results are simplified.
[0060] The second step is to mesh the basic structure that needs topology optimization, establish a finite element model, and select the design area and non-design area.
[0061] The third step is to assign material parameters such as Young's modulus, density, and Poisson's ratio, and apply boundary conditions and loads.
[0062] The fourth step is to determine the optimization parameters of the bidirectional progressive structural optimization method, such as the filtering radius, penalty factor, and maximum number of iterations.
[0063] The fifth step is to perform finite element analysis on the model and calculate the sensitivity under the objective function and constraint conditions. The calculation of sensitivity needs to consider the design-related loads;
[0064] The established optimization model is as follows:
[0065]
[0066] Where, the subscript i represents the unit number, i = 1, 2, 3...N, X is the set of design variables, x i is the structural design variable of the i-th unit, which takes the value of 1 or the set minimum value x min , C(X) is the objective function, i.e. the flexibility of the structure, F T is the transpose of the force matrix, U is the displacement matrix, V * is the target volume, v i is the volume of the i-th unit;
[0067] The design related loads are applied by means of equivalent loads, which are calculated as follows:
[0068] K e u e =f e
[0069] In the formula, Ke represents the element stiffness matrix, u e represents the unit displacement, f e Represents the equivalent nodal forces of the element, and after expanding them:
[0070]
[0071] u e =N e U e
[0072]
[0073] In the formula, K e represents the element stiffness matrix, matrices B and D represent the element geometry matrix and element elastic coefficient matrix respectively, B T represents the transpose of the unit geometry matrix, S e represents the area of the unit, t refers to the thickness of the unit, and u e represents the unit displacement, N e is the unit shape function matrix, U e is the element node displacement, and They represent the equivalent nodal force of the unit volume force and the equivalent nodal force of the unit surface force respectively. The calculation formula for converting the unit volume force into the nodal force is:
[0074]
[0075] In the formula, p v represents the unit volume force vector, N T is the transpose of the shape function, Ω is the unit volume. Taking the plane four-node rectangular unit as an example, the shape function and unit volume force are substituted into the calculation formula of the equivalent nodal force of the unit:
[0076]
[0077] A is the area of the unit, The transpose of the shape function of the ith element is simplified to:
[0078]
[0079] For a planar four-node rectangular element, its shape function can be expressed as:
[0080]
[0081] Where a and b represent half of the length and width of the unit, respectively, and N1(x, y) to N4(x, y) represent the four nodes of each unit.
[0082] Combining the above formulas, we can get the equivalent nodal force of the i-th node of the structural unit under the rotation speed ω:
[0083]
[0084] Where N i is the shape function of the ith element, ρ is the element density, F li is the total centrifugal force of the unit, V represents the total volume of the unit, r i It represents the distance from the geometric center of the i-th unit to the rotation axis. It can be seen from the above formula that for a planar four-node rectangular unit with uniform thickness, the equivalent nodal force under centrifugal load is 1 / 4 of the total centrifugal force of the unit, and the direction is the same as the total centrifugal force of the unit. In the same way, it can be deduced that the equivalent nodal force under gravity load is also 1 / 4 of the total gravity of the unit, and the direction is the same as the total gravity of the unit.
[0085] If the applied load is an external load, the sensitivity calculation formula is:
[0086]
[0087] Wherein, the subscript i represents the serial number of the unit (i = 1, 2, 3 ... N, there are N units in total), α i is the sensitivity of the i-th unit, q is the penalty factor, C is the average flexibility of the structure, x i is the structural design variable of the ith unit, is the structural stiffness matrix of the ith unit, u i is the structural displacement vector of the ith unit, is the transposed matrix of the structural displacement vector of the i-th unit;
[0088] If the applied load is gravity load, centrifugal load and other loads related to the design structure, the sensitivity calculation formula is as follows;
[0089] Centrifugal load:
[0090]
[0091] Gravity load:
[0092]
[0093] Wherein, the subscript i represents the unit number, i = 1, 2, 3...N, α i is the sensitivity of the i-th unit, q is the penalty factor, C is the average flexibility of the structure, x i is the structural design variable of the i-th unit, ω is the applied rotation speed, r i is the rotation radius of the ith unit, v i is the volume of the i-th unit, ρ0 is the initial density, is the force direction matrix, is the structural stiffness matrix of the ith unit, u i is the structural displacement vector of the ith unit, is the transposed matrix of the structural displacement vector of the ith unit, g is the gravitational acceleration, and when the optimized structure does not consider gravity load or centrifugal load, g = 0 or ω = 0, the calculation of sensitivity is the same as that of external load.
[0094] Step 6: Calculate the evolution rate of the current iteration step through the change function and calculate the updated structure volume;
[0095] The evolution rate of the kth iteration is:
[0096]
[0097] Then the volume after the kth iteration is
[0098] Among them, SR k is the evolution rate of the current k-th iteration step, SR max is the maximum evolution rate set, SR min is the minimum evolution rate set, S (i.e. S(y)) is the change function, s1 and s2 are the control parameters set, y is the ratio of the current volume to the original volume (V0 initial volume), v is the set volume constraint, V k is the volume after the kth iteration, V k-1 is the volume after the k-1th iteration, V * is the target volume (V * =V0×v). This function can limit the evolution rate of each superposition between the set maximum evolution rate and the minimum evolution rate, and as the current volume fraction changes, the slope and left and right movement of the function can be controlled by adjusting the values of s1 and s2 respectively.
[0099] Step 7: Update the unit according to the sensitivity and reassign the material properties using the material interpolation model;
[0100] The material interpolation model uses the rational approximate material interpolation model:
[0101]
[0102] Wherein, the subscript i represents the unit number, i = 1, 2, 3...N, ρ(x i ) is the density of the ith unit, x i is the structural design variable of the i-th unit, ρ0 is the initial density, E(x i) is the Young’s modulus of the ith unit, E0 is the initial Young’s modulus, and q is the penalty factor. Generally, a larger penalty factor is conducive to improving the convergence performance of the bidirectional progressive structural optimization method in design-related load problems.
[0103] In the eighth step, the finite element model structure is updated, and it is determined whether the objective function meets the convergence criteria and constraints; if not, return to the fifth step; if satisfied, the optimization is terminated and the model is output; the unit update is carried out by sorting the sensitivity, and the updated structure volume is calculated according to the calculated evolution rate. The units with lower sensitivity are deleted, and the units with higher sensitivity are retained. The constraint condition is that the ratio of the volume of the optimized structure to the volume of the original structure reaches the set volume percentage, and the convergence condition is that the average flexibility change of the structure five times before and after is less than the set value.
[0104] Example 1 Topological Optimization of the Cross-Sectional Structure of a Turbine Disk Subjected to Centrifugal Load
[0105] 1. Select the structure of the aircraft engine turbine disk, because the aerodynamic load, the load generated by the assembly and other loads are less than 10 orders of magnitude different from the centrifugal load. -3 , which has little impact on the optimized structure. Considering the optimization efficiency, these loads are simplified. After analysis, the main load type is centrifugal load.
[0106] 2. Mesh the basic structure to be topologically optimized, and select the design area and the non-design area. In this embodiment, the aspect ratio of the structure is 2:1.
[0107] 3. If Figure 2 As shown in the figure, the structure rotates around a fixed rotation axis at an angular velocity of ω = 1000 rad / s and a rotation radius of r0 = 20 mm. The black area on the right edge of the structure is the non-design domain, which is used to connect the blades on the outer edge of the wheel. The sensitivity is set to 10 in the subsequent calculation. 20 , to prevent it from being deleted during the optimization process, set the material Young's modulus to 210 GPa, the density to 8.45 g / cm, and the Poisson's ratio to 0.3.
[0108] 4. Determine the required parameters of the bidirectional progressive structural optimization algorithm, including filter radius, penalty factor, maximum number of iterations, volume constraint, and convergence judgment. In this example, the optimized filter radius R min , penalty factor q, maximum number of iterations, volume constraint v and convergence judgment e are 2, 5, 200, 40%, and 0.1% respectively.
[0109] 5. Perform finite element analysis on the model and calculate the sensitivity of the objective function and constraints
[0110] The established optimization model is as follows:
[0111]
[0112] Where, the subscript i represents the unit number, i = 1, 2, 3...N, X is the set of design variables, x i is the structural design variable of the i-th unit, which takes the value of 1 or the set minimum value x min , C(X) is the objective function, F T is the transpose of the force matrix, U is the displacement matrix, V * is the target volume, v i is the volume of the i-th unit;
[0113] Since the applied load is a centrifugal load, the sensitivity expression is:
[0114]
[0115] Wherein, the subscript i represents the unit number, i = 1, 2, 3...N, α i is the sensitivity of the i-th unit, q is the penalty factor, C is the average flexibility of the structure, x i is the structural design variable of the i-th unit, ω is the applied rotation speed, r i is the rotation radius of the ith unit, υ i is the volume of the i-th unit, ρ0 is the initial density, is the force direction matrix, is the structural stiffness matrix of the ith unit, u i is the structural displacement vector of the ith unit, is the transposed matrix of the structural displacement vector of the i-th unit.
[0116] 6. Calculate the evolution rate of the current iteration step through the change function, and calculate the updated structure volume;
[0117] The evolution rate of the kth iteration is:
[0118]
[0119] Then the volume after the kth iteration is
[0120] Among them, SR k is the evolution rate of the current k-th iteration step, SR max is the maximum evolution rate set, SR min is the minimum evolution rate set, S(y) is the change function, s1 and s2 are the control parameters set, y is the ratio of the current volume to the original volume, v is the volume constraint set, V k is the volume after the kth iteration, Vk-1 is the volume after the k-1th iteration, V * is the target volume. In this example, s1 and s2 are 20 and -10.
[0121] 7. Update the unit and set x for the inefficient unit i =x min =10 -3 , the remaining units are set to x i =1, and use the material interpolation model to reassign material properties:
[0122]
[0123] Wherein, the subscript i represents the unit number, i = 1, 2, 3...N, ρ(x i ) is the density of the ith unit, x i is the structural design variable of the i-th unit, ρ0 is the initial density, E(x i ) is the Young’s modulus of the ith unit, E0 is the initial Young’s modulus, and q is the penalty factor.
[0124] 8. Update the finite element model structure and determine whether the objective function meets the convergence criteria and constraints. If not, return to step 5; if satisfied, end the optimization and output the model.
[0125] Figure 3 This is the final topology optimization structure obtained by the bidirectional progressive structural optimization method that can be used for aerospace structures. In this example, the constraint condition is the volume constraint. The ratio of the optimized volume to the original volume is 39.9%, which is less than the set 40%. In the five calculations before and after, the average flexibility of the structure changed by 0.07%, which is less than the set 0.1%, and the structure is considered to have converged. Compared with the original structure, the average flexibility of the final optimized structure is reduced by 63%. While the lightweight design is achieved, the overall stiffness of the structure is improved. Compared with the original bidirectional progressive structural optimization method with a fixed evolution rate, the number of iterations for the same structure is reduced from 57 to 32, and the efficiency is improved.
[0126] Figure 4 The figure shows a comparison between the iterative process of the improved bidirectional progressive structural optimization method based on the variation function to determine the evolution rate and the iterative process of the bidirectional progressive structural optimization method with a constant evolution rate. Through the comparison, it can be seen that the method of the present invention can reduce the number of steps required for iteration while ensuring accuracy, thereby improving the efficiency of topology optimization.
[0127] Although the embodiments of the present invention have been disclosed as above, they are not limited to the applications listed in the specification and the embodiments. They can be fully applied to various fields suitable for the present invention. For those familiar with the art, additional modifications can be easily implemented. Therefore, without departing from the general concept defined by the claims and the scope of equivalents, the present invention is not limited to the specific details and the illustrations shown and described herein.
Claims
1. A bidirectional progressive structural optimization algorithm for aerospace structures, characterized in that: The following steps are involved: S1. Select the structure to be optimized and determine the load type; S2. Establish a finite element model and perform meshing; S3, assign material parameters, and apply boundary conditions and loads; S4, determining optimization parameters of the bidirectional progressive structural optimization method; S5. Perform finite element analysis and calculate sensitivity. The sensitivity is calculated under the objective function and constraint conditions, where the objective function is to minimize the average flexibility of the structure and the constraint condition is volume constraint. The established optimization model is as follows: Where, the subscript i represents the unit number, i=1, 2, 3…N, X is the set of design variables, is the structural design variable of the i-th unit, which takes a value of 1 or the minimum value set , C(X) is the objective function, is the transpose of the force matrix, U is the displacement matrix, is the target volume, is the volume of the i-th unit; If the applied load is an external load, the sensitivity calculation formula is: , Wherein, the subscript i represents the unit number, i=1, 2, 3...N, is the sensitivity of the ith unit, q is the penalty factor, C is the average flexibility of the structure, is the structural design variable of the ith unit, is the structural stiffness matrix of the ith unit, is the structural displacement vector of the ith unit, is the transposed matrix of the structural displacement vector of the i-th unit; If the applied load is a centrifugal load, the sensitivity calculation formula is: , in, is the applied speed, is the rotation radius of the ith unit, is the volume of the ith unit, is the initial density, is the direction matrix of the force; If the applied load is a gravity load, the sensitivity calculation formula is: , Where g is the acceleration due to gravity; S6, calculating the evolution rate of the current iteration step and determining the updated structure volume; calculating the evolution rate and the number of units that need to be updated through the change function; The evolution rate of the kth iteration is: , ; Then the volume after the kth iteration is , in, is the evolution rate of the current k-th iteration step, is the maximum evolution rate set, is the minimum evolution rate set, is the change function, s1 and s2 are the set control parameters, y is the ratio of the current volume to the original volume, and v is the set volume constraint. is the volume after the kth iteration, is the volume after the k-1th iteration, is the target volume; S7, update the unit and structure, and reassign material parameters; S8: Determine whether the convergence criteria and constraints are met. If not, return to step S5; if satisfied, end the optimization and output the model.
2. The bidirectional progressive structural optimization algorithm applicable to aerospace structures according to claim 1, characterized in that: In step S1, the load types include design-related loads and external loads, and the design-related loads include gravity loads and centrifugal loads; When determining the load type, ignore loads that differ greatly in magnitude and have little effect on the results.
3. The bidirectional progressive structural optimization algorithm applicable to aerospace structures according to claim 1, characterized in that: In the step S2, the design area and the non-design area are selected.
4. The bidirectional progressive structural optimization algorithm applicable to aerospace structures according to claim 1, characterized in that: In step S3, the material parameters include Young's modulus, density and Poisson's ratio; In step S4, the optimization parameters include filter radius, penalty factor, maximum number of iteration steps, volume constraint and convergence judgment.
5. The bidirectional progressive structural optimization algorithm applicable to aerospace structures according to claim 1, characterized in that: In step S7, the material parameters are reassigned using a material interpolation model, and the interpolation model is as follows: , Wherein, the subscript i represents the unit number, i=1, 2, 3...N, is the density of the ith unit, is the structural design variable of the ith unit, is the initial density, is the Young's modulus of the ith unit, is the initial Young's modulus, and q is the penalty factor.
6. The bidirectional progressive structural optimization algorithm applicable to aerospace structures according to claim 1, characterized in that: In step S8, it is determined whether the objective function meets the convergence criteria and constraint conditions. The objective function is to minimize the average flexibility of the structure, the constraint condition is the volume constraint, and the convergence criterion is that the average flexibility change of the structure five times before and after is not greater than the set value, as shown in the following formula: Where C is the average flexibility of the structure, the subscript represents the number of iterations, k represents the kth iteration, j is the count, and e is the set value of the convergence criterion.
