A parameter optimization design method for a composite material main box section

Abstract

The application provides a composite material main box section parameter optimization design method, including the following steps: screening the main design working conditions of the strength constraint and the stiffness constraint of the composite material main box section; taking the number of plies as an optimization variable, randomly generating each optimization variable, and equally dividing the initial ply thickness corresponding to the optimization variable in the form of a super layer according to the initial ply sequence to obtain the initial ply thickness of each optimization variable; optimizing the initial ply thickness of each optimization variable to obtain a basic ply thickness; obtaining a ply library of the composite material main box section, the ply library containing a ply sequence corresponding to the ply thickness; judging whether the finite element model of the composite material main box section meets the strength constraint and the stiffness constraint requirement; if not, performing full-strength parameter fine tuning until the constraint requirement is met; and performing thickness process constraint checking on the composite material main box section to ensure that the thickness step difference of adjacent thickness partitions meets the design requirement.

Description

TECHNICAL FIELD

[0001] The application belongs to the field of strength calculation, and particularly relates to a parameter optimization design method for a composite main box section. BACKGROUND

[0002] There are many difficulties in the optimization design of the main structure of a composite structure, such as many optimization variables, complex strength design criteria, and severe process constraints. In recent years, many domestic and foreign scholars have carried out in-depth research on how to improve the solving efficiency and accuracy of the difference method and the semi-analytical method. As for the analytical method itself, the analytical method is recognized as the highest in efficiency and accuracy among all sensitivity calculation methods, but it has not been widely studied and applied due to its complex theoretical derivation. Generally, the analytical optimization algorithm needs to assemble the differential stiffness matrix of the structure through the differential stiffness matrix of the element, but the differential stiffness matrix is generally difficult to derive analytically, so many commercial software avoid the analytical method and use the difference method or the semi-analytical method to solve the sensitivity.

[0003] The traditional method has perfect mathematical theory, fast convergence speed and other advantages, but also has significant disadvantages: the algorithm is relatively complex, the universality is poor, only for some specific problems, and has strict requirements for the search space; the above methods are mostly local optimization methods, and the optimization results of the algorithm are closely related to the selection of the initial point. With the development of science and technology, the problems encountered in practical applications are becoming more and more complex, many of which cannot be expressed by functions, and the traditional method cannot obtain satisfactory results. Therefore, in recent decades, various stochastic algorithms have been proposed, such as evolutionary algorithm, genetic algorithm, particle swarm algorithm, ant colony algorithm, immune algorithm and the like. This kind of algorithm is based on group heuristic optimization algorithm, and its idea comes from the behavior of nature. Compared with the traditional method, this kind of method is a “black box” mechanism, and the algorithm has good universality, but the convergence speed is slow. The calculation efficiency, convergence and rationality of the optimization result are far from the engineering application.

[0004] Therefore, it is desirable to have a technical solution to overcome or at least alleviate at least one of the aforementioned deficiencies of the prior art. SUMMARY

[0005] The purpose of the present application is to provide a parameter optimization design method for a composite main box section to solve at least one problem existing in the prior art.

[0006] The technical solution of the present application is:

[0007] A parameter optimization design method for a composite main box section, comprising:

[0008] Step one, through finite element calculation and analysis evaluation, the main design working conditions of the strength constraints and the stiffness constraints of the composite main box section are screened out;

[0009] Step two, taking the number of plies as the optimization variable, randomly generating each optimization variable, and dividing the initial ply thickness corresponding to the optimization variable into super layers according to the initial ply sequence to obtain the initial ply thickness of each optimization variable;

[0010] Step three, using a two-level genetic algorithm to optimize the initial ply thickness of each optimization variable to obtain the basic ply thickness;

[0011] Step four, obtaining the ply library of the composite main box section, which contains the ply sequence corresponding to the ply thickness;

[0012] Step five, according to the basic ply thickness, calling the corresponding ply sequence from the ply library to construct the finite element model of the composite main box section, and judging whether the finite element model of the composite main box section meets the strength constraint and stiffness constraint requirements;

[0013] If not, perform full strength parameter fine tuning until the constraint requirements are met;

[0014] Step six, according to the process constraint of the composite structure design, performing thickness process constraint check on the composite main box section to ensure that the thickness step difference of adjacent thickness partitions meets the design requirements.

[0015] In at least one embodiment of the present application, in step two, the initial ply sequence is 45° / 90° / -45° / 0° / 0° / -45° / 90° / 45°.

[0016] In at least one embodiment of the present application, in step three, using a two-level genetic algorithm to optimize the initial ply thickness of each optimization variable to obtain the basic ply thickness, including:

[0017] 3.1, first-level genetic algorithm optimization is performed;

[0018] 3.11, in the first-level genetic algorithm optimization, the fitness f of the jth optimization variable of the ith population is determined ij Value;

[0019] 3.12, the first-level genetic algorithm optimization uses single-point crossover method, randomly sets a crossover point in the individual code string, and then exchanges part of the code operators of two paired individuals at the point;

[0020] 3.13, the first-level genetic algorithm optimization uses basic bit mutation method, randomly specifies the value of a certain gene locus in the individual code string with a certain mutation probability to perform mutation operation;

[0021] 3.2, second-level genetic algorithm optimization is performed;

[0022] 3.21、In the second level genetic algorithm optimization, determine the overall fitness s of the ith population i ;

[0023] 3.22、The second level genetic algorithm optimization adopts arithmetic crossover method, and a new individual is generated through linear combination of two individuals;

[0024] 3.23、In the second level genetic algorithm optimization, the crossover probability is dynamically adjusted according to the state of evolution;

[0025] 3.24、The second level genetic algorithm optimization adopts real number mutation method, and a plurality of positions are randomly generated, and the unit individual parameters at the corresponding positions are randomly taken as numerical values in the parameter boundary range;

[0026] 3.3、After the two-level genetic algorithm optimization is completed, the preferred parent elite sample is reserved.

[0027] In at least one embodiment of the present application, in step 3.11, the jth optimization variable fitness f of the ith population ij is taken as:

[0028]

[0029] Wherein, x ij represents the safety margin of the jth unit individual of the ith population, s i is the global deformation constraint fitness value of the ith population, K1 and K2 are weighting coefficients.

[0030] In at least one embodiment of the present application, in step 3.21, the overall fitness s of the ith population i is taken as:

[0031]

[0032] Wherein, S is the extreme value of the displacement of the whole machine, S max and S min are the upper and lower limits of the displacement constraint, and K3 and K4 are weighting coefficients.

[0033] In at least one embodiment of the present application, in step 3.22, the new individual generated through linear combination of two individuals includes:

[0034] Two crossover point positions are randomly generated, and then arithmetic crossover is performed between the two populations and , and the new individual generated is:

[0035]

[0036]

[0037] wherein, is a preferred parent, is a population of offspring with poor fitness, and a is a random weight coefficient.

[0038] In at least one embodiment of the present application, in step 3.23, the crossover probability is dynamically adjusted according to the state of evolution as follows:

[0039]

[0040] wherein, f i,max is the maximum fitness value of the population, and f i,ave is the average fitness value of the population.

[0041] In at least one embodiment of the present application, in step 3.3, the following strategy is used to retain the elite sample of the preferred parent:

[0042] Given two decision variables X W and X s represent the weight and stiffness constraints, respectively;

[0043] If and only if , if f i (X s ) all satisfy the constraint conditions, then sort according to the weight X W to find the optimal population;

[0044] If and only if , if at least one f i (X s ) does not satisfy the constraint condition, and at least one f i (X W ) satisfies the constraint condition, then sort according to the weight X W in the population that satisfies the constraint to find the optimal population;

[0045] If and only if , if there is no f i (X s ) that satisfies the constraint condition, then sort according to f i (X s ), that is, sort according to the program that approaches the stiffness constraint target.

[0046] In at least one embodiment of the present application, in step four, the ply library of the main box section of the composite material is obtained, comprising:

[0047] 4.1, according to the composite wallboard, wing beam composite material main box section structure partition, in each structure partition, according to the volume weight coefficient of different thickness partition, the weight coefficient of the largest layer thickness is screened out, and each layer thickness corresponds to only one layer proportion;

[0048] 4.2, the layer proportion of each structure partition is counted, and the layer library corresponding to the layer proportion and the layer thickness of the composite wallboard and the wing beam is established.

[0049] In at least one embodiment of the application, in step five, the parameter fine tuning strategy of full strength parameter fine tuning is:

[0050]

[0051] Wherein, C i,j The margin of the jth strength control criterion of the ith variable, M lower , M upper The lower limit and upper limit of the minimum margin of variable strength, Margin is the minimum margin under all strength constraint controls, Respectively represent the value of the ith variable before and after parameter adjustment.

[0052] In at least one embodiment of the application, in step six, according to the process constraint of composite structure design, the thickness process constraint check of composite main box section is carried out, so as to ensure that the thickness step difference of adjacent thickness partitions meets the design requirements, including:

[0053] 6.1, according to the position relationship of each structure partition, each thickness partition of the composite main box section is coded in plane;

[0054] 6.2, the thickness step difference of each partition is checked from the maximum thickness of the structure, if the thickness step difference between the current coded thickness partition and the adjacent thickness partition exceeds the step difference critical value, the thickness monotonic gradient direction of the adjacent thickness partition is calculated;

[0055] 6.3, according to the thickness monotonic gradient direction, the current coded thickness partition is incremented by one in the layer library, until the thickness step difference meets the design requirements.

[0056] In at least one embodiment of the application, it further includes step seven, strength calculation check of the composite main box section is carried out, and the strength margin under the current strength constraint condition is output.

[0057] The application has at least the following beneficial technical effects:

[0058] The composite main box section parameter optimization design method of the application can meet the optimization design requirements of the composite main box section structure layer proportion, layer thickness and the like under the strength constraint conditions of composite strain and stability, and can realize engineering design of the composite overall structure meeting the process constraint in combination with the layer library, greatly improving the strength design efficiency of the composite main box section structure and shortening the development cycle. BRIEF DESCRIPTION OF DRAWINGS

[0059] Figure 1 is a binary encoding cross schematic diagram of an embodiment of the application;

[0060] Figure 2 is a binary mutation schematic diagram of an embodiment of the application;

[0061] Figure 3 is a composite main box section wallboard partition coding processing schematic diagram of an embodiment of the application. DETAILED DESCRIPTION

[0062] In order to make the purpose, technical scheme and advantages of the application clearer, the technical scheme of the embodiments of the application will be described in more detail below in combination with the drawings of the embodiments of the application. In the drawings, the same or similar notations represent the same or similar elements or elements with the same or similar functions throughout. The described embodiments are part of the embodiments of the application, not all of the embodiments. The embodiments described below by reference to the drawings are exemplary and are intended to explain the application, and cannot be understood as a limitation of the application. Based on the embodiments in the application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the application. The embodiments of the application will be described in detail below in combination with the drawings.

[0063] In the description of the application, it should be understood that the orientations or positional relationships indicated by the terms "center", "longitudinal", "transverse", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer" and the like are based on the orientations or positional relationships shown in the drawings, and are only for the convenience of describing the application and simplifying the description, and therefore cannot be understood as indicating or implying that the devices or elements referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the scope of protection of the application.

[0064] The embodiments of the application will be described in detail below in combination with the drawings. Figures 1 to 3 The application will be described in further detail.

[0065] The application provides a composite main box section parameter optimization design method, comprising the following steps:

[0066] Step one, through finite element calculation and analysis evaluation, the main design conditions of strength constraint and stiffness constraint of the composite main box section are screened out;

[0067] Step two, taking the number of plies as the optimization variable, each optimization variable is randomly generated, and the initial ply thickness corresponding to the optimization variable is equally divided in the form of super layer according to the initial ply sequence to obtain the initial ply thickness of each optimization variable;

[0068] Step three, a two-level genetic algorithm is used to optimize the initial ply thickness of each optimization variable to obtain the basic ply thickness;

[0069] Step four, the ply library of the composite main box section is obtained, and the ply sequence corresponding to the ply thickness is included in the ply library;

[0070] Step five, the corresponding ply sequence is called from the ply library according to the basic ply thickness to construct the finite element model of the composite main box section, and it is judged whether the finite element model of the composite main box section meets the strength constraint and stiffness constraint requirements;

[0071] If not, the full strength parameter fine tuning is performed until the constraint requirements are met;

[0072] Step six, according to the process constraint of the composite structure design, the thickness process constraint check is performed on the composite main box section to ensure that the thickness step difference of adjacent thickness partitions meets the design requirements.

[0073] The composite main box section parameter optimization design method of the application, first, in step one, the preliminary finite element calculation and analysis evaluation is performed, and the main design conditions of strength constraint and stiffness constraint are screened out. In step two, taking the number of structure grid elements as the optimization variable, each optimization variable is randomly generated, and the variable corresponding ply thickness is equally divided in the form of super layer according to the initial ply sequence of 45° / 90° / -45° / 0° / 0° / -45° / 90° / 45°, and the corresponding stiffness attribute file is generated.

[0074] In step three of the composite main box section parameter optimization design method of the application, a two-level genetic algorithm is used to perform fitness evaluation, selection, encoding and decoding, crossover and mutation from two levels of local element individuals and global whole variable set; and according to the fitness evaluation, a memory library is established to retain the elite group until the optimization converges, and the basic ply thickness of the composite main box section structure is optimized.

[0075] Specifically, in step three of the embodiment, a two-level genetic algorithm is used to optimize the initial ply thickness of each optimization variable to obtain the basic ply thickness, including:

[0076] 3.1, first-level genetic algorithm optimization is performed;

[0077] 3.11, in the first genetic algorithm optimization, determine the jth optimization variable fitness f of the ith population ij value;

[0078] In step 3.11, the jth optimization variable fitness f of the ith population ij value:

[0079]

[0080] Where x ij represents the safety margin of the jth unit individual of the ith population, s i is the fitness value of the global deformation constraint of the ith population, K1 and K2 are weighting coefficients.

[0081] 3.12, the first genetic algorithm optimization uses single-point crossover method, randomly sets a crossover point in the individual code string, and then exchanges part of the code operators of two paired individuals at the point, as shown in Figure 1 ;

[0082] 3.13, the first genetic algorithm optimization uses basic bit mutation method, randomly specifies the value of a certain gene locus in the individual code string with a certain mutation probability, and performs mutation operation, as shown in Figure 2 ;

[0083] 3.2, second genetic algorithm optimization is carried out;

[0084] 3.21, in the second genetic algorithm optimization, determine the overall fitness s i value;

[0085] In step 3.21, the overall fitness s i value:

[0086]

[0087] Where S is the extreme value of the whole machine displacement (wing tip deformation, torsion angle, etc.), S max and S min are the upper and lower limits of displacement constraint, K3 and K4 are weighting coefficients.

[0088] 3.22, the second genetic algorithm optimization uses arithmetic crossover method, and a new individual is generated by linear combination of two individuals;

[0089] Real number coding crossover is also called real value recombination, which mainly includes discrete recombination, intermediate recombination and arithmetic recombination.

[0090] In step 3.22, a new individual is generated by linear combination of two individuals, including:

[0091] Randomly generate two crossover positions, then perform arithmetic crossover between two populations and to produce new individuals:

[0092]

[0093] wherein, is the preferred parent population, is the poor child population, and a is a random weight coefficient.

[0094] This allows the poor child population to inherit the site genes of the preferred parent population. In actual optimization, the weight coefficient of the preferred parent population can be appropriately increased.

[0095] 3.23, in the second genetic algorithm optimization, the crossover probability is dynamically adjusted according to the state of evolution;

[0096] In order to make the algorithm search to the global optimum, in step 3.23, the crossover probability is dynamically adjusted according to the state of evolution:

[0097]

[0098] wherein, f i,max is the maximum fitness value of the population, and f i,ave is the average fitness value of the population.

[0099] 3.24, the second genetic algorithm optimization adopts real number mutation method, randomly selects multiple positions, and randomly selects the value of the unit individual parameter in the corresponding position within the parameter boundary range;

[0100] 3.3, after the two-level genetic algorithm optimization, the preferred parent elite sample is retained.

[0101] In step 3.3, the following strategy is adopted to retain the preferred parent elite sample:

[0102] Given two decision variables X W and X s , representing weight and stiffness constraints, respectively;

[0103] If and only if , if f i (X s ) satisfies the constraint condition, then according to the weight X W , the optimal population is sorted out;

[0104] If and only if , if at least one f i (X s ) does not satisfy the constraint condition, and at least one fi (X W ) meet the constraint condition, then sort the population meeting the constraint according to the weight X W , find the optimal population;

[0105] iff , if there is no f i (X s ) meet the constraint condition, then sort according to f i (X s ), that is, sort according to the program of approaching the stiffness constraint target.

[0106] In each optimization iteration step, a small number of optimal K "elite" populations are selected for preservation. During the optimization process, if a small number of populations do not meet the stiffness requirement, the elite preserved from the previous optimization iteration step is randomly replaced with the population with poor fitness; if the number of populations meeting the global stiffness constraint is less than the parent population number (P), then M (M < K) populations are randomly selected from the elite population other than the previous optimization step to randomly replace the existing population.

[0107] The composite main box section parameter optimization design method of the application, in step four, the ply library of the composite main box section is obtained, comprising:

[0108] 4.1, according to the composite wall plate and wing beam, the structure of the composite main box section is divided, in each structure partition, according to the volume weight coefficient of different thickness partitions, the ply proportion with the largest weight coefficient of ply thickness is screened out, so as to ensure that each ply thickness corresponds to only one ply proportion;

[0109] 4.2, the ply proportion of each structure partition is counted, and the ply library corresponding to the ply proportion and the ply thickness of the composite wall plate and the wing beam is established respectively.

[0110] In step five of the composite main box section parameter optimization design method of the application, according to the initial optimization thickness, the ply sequence table corresponding to the thickness is called from the ply library, and the full strength design is used to measure the fine tuning parameter until the strength convergence requirement is met or the specified optimization iteration step is reached.

[0111] According to the optimization thickness, the ply sequence table corresponding to the thickness is selected from the ply library, and if there is no ply thickness corresponding to the thickness, the ply sequence table corresponding to the adjacent upper level thickness is selected.

[0112] After updating the model, full strength parameter fine tuning is performed until the strength constraint requirement is met. The parameter fine tuning strategy is as follows:

[0113]

[0114] Wherein, C i,jMargin of the ith variable under the jth intensity control criterion, M lower , M upper are the lower and upper limits of the minimum margin of variable intensity, respectively, and Margin is the minimum margin under all intensity constraints, respectively, represent the values of the ith variable before and after adjustment.

[0115] After optimization meets the convergence condition, parameter adjustment is stopped.

[0116] In step six of the composite main box section parameter optimization design method of the application, according to the process constraints of composite structure design, thickness process constraint check is performed on the composite main box section to ensure that the thickness step difference of adjacent thickness partitions meets the design requirements, including:

[0117] 6.1, according to the positional relationship of each structural partition, each thickness partition of the composite main box section is coded in the plane;

[0118] 6.2, each partition is checked for thickness step difference starting from the maximum thickness of the structure, and if the thickness step difference between the current coded thickness partition and the adjacent thickness partition exceeds the step difference threshold, the thickness monotonic gradient direction of the adjacent thickness partition is calculated;

[0119] 6.3, according to the thickness monotonic gradient direction, the current coded thickness partition is incremented by one in the thickness in the ply library until the thickness step difference meets the design requirements.

[0120] The thickness in the ply library is divided into multiple thickness steps from high to low, and thickness increment means incrementing by one to a thicker thickness step.

[0121] In the preferred embodiment of the application, each thickness partition of the whole composite structure is coded in the plane according to the positional relationship of each structural partition, as shown in the following formula and Figure 3 .

[0122]

[0123] Each partition is checked for thickness step difference starting from the maximum thickness of the structure, and if A i,j and its adjacent structural thickness exceeds the step difference threshold AT, such as A i-1,j -A i,j ≥ AT, the gradient change direction rule of the thickness of the adjacent structural unit is calculated first, as shown in Figure 3 ; according to the calculated thickness monotonic gradient direction, A i,j is incremented by one in the thickness in the ply library until A i-1,j -A i,j ≤ AT, A i,j-1 -A i,j ≤ AT, Ai-1,j-1 -A i,j ≤ ΔT.

[0124] The composite main box section parameter optimization design method of the present application further comprises the following step seven after the process inspection of each variable thickness is completed: performing strength calculation inspection on the composite main box section, and outputting the strength margin under the current strength constraint condition.

[0125] The composite main box section parameter optimization design method of the present application can solve the optimization problem of the composite main box section structure with multiple variables, complex constraints and strict process requirements. It can meet the optimization design requirements of the composite main box section structure, such as the proportion and thickness of the ply, under the strength constraint conditions such as the strain and stability of the composite material, and can realize the engineering design of the composite overall structure that meets the process constraints in combination with the ply library. The strength design efficiency of the composite main box section structure is greatly improved, and the development cycle is shortened.

[0126] The above is only a specific embodiment of the present application, but the protection scope of the present application is not limited thereto. Any changes or replacements that can be easily thought of by those skilled in the art within the technical scope disclosed in the present application should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. A method of parametric optimization design of a composite material main box section, characterized in that, Comprise: Step one, through finite element calculation and analysis evaluation, screening out the strength constraint and stiffness constraint of the main box section of the composite material; Step two, taking the number of plies as the optimization variable, randomly generating each optimization variable, and equally dividing the initial ply thickness corresponding to the optimization variable in the form of super layer according to the initial ply sequence to obtain the initial ply thickness of each optimization variable; Step three, using two-stage genetic algorithm, the initial ply thickness of each optimization variable is optimized to obtain the basic ply thickness; Step four, obtaining the ply library of the main box section of the composite material, the ply library contains the ply sequence corresponding to the ply thickness; Step five, according to the basic ply thickness, the corresponding ply sequence is called from the ply library to construct the finite element model of the main box section of the composite material, and whether the finite element model of the main box section of the composite material meets the strength constraint and stiffness constraint requirements is judged; If not, the full strength parameter fine tuning is performed until the constraint requirements are met; Step six, according to the process constraint of the composite structure design, the thickness process constraint check is performed on the main box section of the composite material to ensure that the thickness step difference of adjacent thickness partitions meets the design requirements; In step three, the two-stage genetic algorithm is used to optimize the initial ply thickness of each optimization variable to obtain the basic ply thickness, which comprises: 3.1, first-stage genetic algorithm optimization is performed; 3.

11. In the first level genetic algorithm optimization, determine the fitness f of the jth optimization variable of the ith population ij values; 3.12, the first-stage genetic algorithm optimization adopts single-point crossover method, a crossover point is randomly set in the individual code string, and then the part of the code operator of two paired individuals is exchanged at the point; 3.13, the first-stage genetic algorithm optimization adopts basic bit mutation method, a certain mutation probability is randomly specified for the value of a certain gene locus in the individual code string to perform mutation operation; 3.2, second-stage genetic algorithm optimization is performed; 3.21 In the second level genetic algorithm optimization, determine the overall fitness s of the i-th population i values; 3.22, the second-stage genetic algorithm optimization adopts arithmetic crossover method, a new individual is generated through linear combination of two individuals; 3.23, in the second-stage genetic algorithm optimization, the crossover probability is dynamically adjusted according to the evolution state; 3.24, the second-stage genetic algorithm optimization adopts real number mutation method, a plurality of positions are randomly selected, and the unit individual parameters at the corresponding positions are randomly taken as values within the parameter boundary range; 3.3, after the two-stage genetic algorithm optimization is completed, the preferred parent elite samples are retained.

2. The method of claim 1, wherein, In step two, the initial ply sequence is 45° / 90° / -45° / 0° / 0° / -45° / 90° / 45°.

3. The method of claim 1, wherein, In step 3.11, the fitness f of the jth optimization variable of the ith population ij takes the value: wherein x ij represents the safety margin of the jth unit individual of the ith population, s i is the fitness value of the global deformation constraint of the ith population, and K1 and K2 are weighting coefficients.

4. The method of claim 3, wherein, In step 3.21, the overall fitness s of the ith population i takes the value: Wherein, S is the full machine displacement extreme value, S max And S min Is the displacement constraint upper and lower limit, K3, K4 is the weighting coefficient.

5. The method of claim 4, wherein, In step 3.22, the new individual generated through linear combination of two individuals comprises: Two crossover point positions are randomly generated and then arithmetic crossover is performed between two populations with the resulting new individuals being: wherein, is a preferred parent, is a population of offspring with poor fitness, and a is a random weighting factor.

6. The method of claim 5, wherein, In step 3.23, the crossover probability is dynamically adjusted according to the evolution state as follows: where f i,max is the maximum fitness value of the population, f i,ave is the average fitness value of the population.

7. The method of claim 6, wherein, In step 3.3, the preferred parent elite samples are retained by using the following strategy: Given two decision variables X W and X s , respectively, represent weight and stiffness constraints; If and only if f i (X s ) all satisfy the constraint condition, sort by weight X W to find the optimal population; iff there exists at least one f i (X s ) does not satisfy the constraint condition, and there also exists at least one f i (X W ) satisfies the constraint condition, then the optimal population is found by sorting the populations satisfying the constraint according to the weight X W ​ iff there is no f i (X s ) satisfying the constraints, then sort according to f i (X s ), i.e. according to the procedure that approximates the stiffness constraint target.

8. The method of claim 7, wherein, In step four, the ply library of the main box section of the composite material is obtained, which comprises: 4.1, according to the composite wall plate and wing beam, the main box section of the composite material is structurally partitioned, in each structural partition, according to the volume weight coefficient of different thickness partitions, the ply proportion with the largest weight coefficient of ply thickness is screened out to ensure that each ply thickness corresponds to only one ply proportion; 4.2, Count the ply proportion of each structural partition, and establish the ply library corresponding to the ply proportion and ply thickness of the composite panel and spar.

9. The method of claim 8, wherein, In step five, the parameter fine-tuning strategy for full strength parameter fine-tuning is: where C i,j is the margin of the jth intensity control criterion for the ith variable, M lower , M upper are the lower and upper limits of the minimum margin of variable intensity, Margin is the minimum margin of all intensity control, respectively represent the value of the ith variable before and after the parameter adjustment.

10. The method of claim 9, wherein, In step six, according to the process constraints of composite structure design, the thickness process constraint check of the composite main box section is carried out to ensure that the thickness step difference of adjacent thickness partitions meets the design requirements, including: 6.1, According to the positional relationship of each structural partition, each thickness partition of the composite main box section is coded in the plane; 6.2, The step difference of each partition is checked from the maximum thickness of the structure, and if the thickness step difference between the current coded thickness partition and the adjacent thickness partition exceeds the critical value, the thickness monotonic gradient direction of the adjacent thickness partition is calculated; 6.3, According to the thickness monotonic gradient direction, the current coded thickness partition is incremented by one level of thickness in the ply library until the thickness step difference meets the design requirements.

11. The method of claim 10, wherein, It also includes step seven, strength calculation check of the composite main box section, and output of the strength margin under the current strength constraint condition.

Images