CN110728081A - Composite material layering sequence optimization system - Google Patents
Composite material layering sequence optimization system Download PDFInfo
- Publication number
- CN110728081A CN110728081A CN201810699919.2A CN201810699919A CN110728081A CN 110728081 A CN110728081 A CN 110728081A CN 201810699919 A CN201810699919 A CN 201810699919A CN 110728081 A CN110728081 A CN 110728081A
- Authority
- CN
- China
- Prior art keywords
- optimization
- layering
- dimensional unit
- composite material
- module
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Landscapes
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to the technical field of computer simulation, in particular to a composite material layering sequence optimization system which comprises a database manufacturing module, a composite material strength checking module, a size definition optimization module, a size optimization algorithm module, a layering library automatic optimization module and a result visualization module. The method refines the traditional strength dimension definition process of the aeronautical structure into an automatic optimization solver, fully considers the completeness and the engineering of the composite material structure dimension definition, and provides a feasible solution with engineering significance for large-scale composite material structure dimension optimization design.
Description
Technical Field
The invention relates to the technical field of computer simulation, in particular to a composite material layering sequence optimization system.
Background
The composite material has the advantages of light weight, high specific strength, high specific rigidity, strong designability, good fatigue fracture resistance, corrosion resistance, good dimensional stability, convenience for large-area integral forming and the like, saves weight by 40 percent under the condition of the same structure compared with a metal structure, greatly improves the fuel efficiency and improves the integral performance of an airplane. Composite structures are increasingly being used on aerospace vehicles because of the above advantages.
The large number of applications for composite structures requires increased knowledge of the composite structure design to avoid some low-level design errors. The optimization design problem that the optimization of the structure size of the existing composite material faces, two of them are particularly outstanding: firstly, the size optimization design scale of the composite material is too large, particularly for optimization design variables; and secondly, optimizing the design of the composite material structure layering.
Disclosure of Invention
In view of the above disadvantages of the prior art, an object of the present invention is to provide a system for optimizing a layering sequence of a composite material, which solves the problem of large-scale optimization design of a composite material structure, and simultaneously fully considers the completeness and engineering of the size definition of the composite material structure, thereby greatly reducing the optimization scale.
The embodiment of the invention provides a composite material layering sequence optimizing system, which comprises:
the database manufacturing module is used for extracting the original model data according to the created generalized finite element GFEM model, manufacturing the original model data into a database and providing input data for the system;
the composite material strength checking module is used for checking the static strength, the stability and the connection strength of the composite material based on the finite element model;
the size definition optimization module is used for optimizing the size definition of the model according to the user target safety margin and finishing the strength checking iteration;
the size optimization algorithm module is used for converting the constrained optimization problem into an unconstrained optimization problem by adopting a multiplier penalty function method, and solving an extreme value of the unconstrained optimization problem by adopting a BFGS quasi-Newton method;
the automatic layering library optimization module adopts an immune genetic algorithm, combines a composite material structure and a process layering design criterion, performs layering sequence optimization design and automatically generates a layering library;
and the result visualization module checks the intensity checking analysis result through a safety margin cloud picture, a result label or a result table.
Further, in the above system, the database includes a ply information database, a load information database, and a material information database.
Further, in the above system, the composite static strength verification includes laminate tensile verification, compressive and shear strain level verification.
Further, in the system, the composite material stability check comprises a one-dimensional unit local instability and pressure loss check, a stiffened plate column buckling check, and a two-dimensional unit compression, shearing and bending instability check.
Further, in the above system, the size definition optimization procedure is as follows:
1) importing an optimization model;
2) calculating the optimized area according to the specified working condition and the specified constraint;
3) optimizing structural units which do not meet the constraint in the optimized region;
4) if the optimized regions all meet the constraint but do not reach the optimization index (tolerance and iteration times), selecting the structural unit with the maximum constraint value in the optimized regions for optimization until the optimized regions meet the optimization index;
5) exporting the model meeting the optimization index, and submitting the model to a solver Nastran for calculation;
6) updating an optimization model load database;
7) calculating the optimized area according to the specified working condition and the specified constraint, and ending the process if the optimized area meets the optimized indexes; if not, the operation is returned to the step 2 for operation.
Further, in the above system, the calculating the optimized region according to the specified operating condition and the specified constraint includes: one-dimensional unit working condition circulation and two-dimensional unit working condition circulation;
the one-dimensional unit working condition cycle comprises the following steps:
extracting load information, layering information and allowable value information corresponding to the one-dimensional unit;
performing one-dimensional unit load judgment and rib plate judgment after compression according to one-dimensional unit static strength check, rib plate column buckling, one-dimensional unit local stability and one-dimensional unit pressure loss;
the two-dimensional unit working condition cycle comprises the following steps:
extracting load information, layering information and allowable value information corresponding to the two-dimensional unit;
and judging the two-dimensional unit load according to the static strength check of the two-dimensional unit, the compression-shear combination of the two-dimensional unit and the bending-shear combination of the two-dimensional unit.
Further, in the above system, the algorithm flow for converting the constrained optimization problem into the unconstrained optimization problem by the size optimization algorithm module using the multiplier penalty function method is as follows:
1) selecting an initial point X0∈RnInitial Lagrange coefficient (. mu.)i)1∈Rl、(λi)1∈RmInitial penalty factor σ1Greater than 0, tolerance 0 < epsilon-1, contrast ratio
The penalty factor increment coefficient eta is more than 1, and k is equal to 1;
2) with Xk-1For the initial point, the method of BFGS is adopted to solve min psi (X, (. mu.) (μ)i)k，(λi)k，σk) Has an extreme value of Xk；
3) Computing
If beta iskStopping iteration and outputting X when epsilon is less than or equal tokCalculating a minimum value f (X) for an approximate minimum point of the primitive functionk) Otherwise, executing step 4;
4) update the penalty factor if
Let sigmak+1＝ησkElse σk+1＝σkAnd updating the Lagrange coefficient, and calculating according to the following formula:
(μi)k+1＝(μi)k-σk+1hi(Xk)
(λi)k+1＝max{0,(λi)k-gi(Xk)}
5) and k is k +1, and the operation of the step 2 is returned.
Further, in the above system, the size optimization algorithm module adopts a BFGS quasi-newton method to solve the extreme value as follows:
1) given parameter deltaE (0,1), σ e (0,1), initial point X0∈RnInitial positive definite matrix B0(taking the identity matrix InOr G (X)0) 0 < epsilon-1, let k be 0;
2) calculate gk＝▽f(Xk) If | gkLess than or equal to epsilon, convergence and X outputkCalculating a minimum value f (X) for an approximate minimum point of the primitive functionk) Otherwise, executing step 3;
3) solving a system of linear equations Bkd＝-gkTo obtain a solution dkLet m bekIs the smallest non-negative integer m that satisfies the following inequality:
order to
Xk+1＝Xk+αkdkB is determined by the BFGS correction formulak+1；
4) And k is k +1, and the operation of the step 2 is returned.
Further, in the above system, the BFGS correction formula
The following were used:
wherein a displacement sk＝xk+1-xkDifference of displacement yk＝gk+1-gk。
Further, in the system, the process of performing layering sequence optimization design by the layering library automatic optimization module through an immune genetic algorithm is as follows:
1) gene coding, namely converting the layering sequence into a chromosome string according to the layering sequence theta i, wherein 1 is 1, 2, 3.n.n, and n is the number of layering, wherein the layering angle coding length of each layer is n, integer coding is adopted, the angle corresponding to the layering coding is fixed, and the optimized layering can obtain the corresponding angle after decoding;
2) fitness analysis, which adjusts the gene string according to constraints, and by chromosome string adjustment, no invalid chromosomes are generated, so that the objective function value is directly used as the fitness function value, namely:
f＝Nxcr
3) genetic operation, wherein the objective function is antigen, the layering sequence is antibody, the population selection probability is determined according to the promotion and inhibition effect between the antibody and the antigen, and then the selection, crossing and mutation operations are carried out.
Compared with the prior art, the composite material layering sequence optimization system comprises a database manufacturing module, wherein the database manufacturing module is used for extracting the original model data according to the created generalized finite element GFEM model to manufacture a database and provide input data for the system; the composite material strength checking module is used for checking the static strength, the stability and the connection strength of the composite material based on the finite element model; the size definition optimization module is used for optimizing the size definition of the model according to the user target safety margin and finishing the strength checking iteration; the size optimization algorithm module is used for converting the constrained optimization problem into an unconstrained optimization problem by adopting a multiplier penalty function method, and solving an extreme value of the unconstrained optimization problem by adopting a BFGS quasi-Newton method; the automatic layering library optimization module adopts an immune genetic algorithm, combines a composite material structure and a process layering design criterion, performs layering sequence optimization design and automatically generates a layering library; and the result visualization module checks the intensity checking analysis result through a safety margin cloud picture, a result label or a result table. The method refines the traditional strength dimension definition process of the aeronautical structure into an automatic optimization solver, fully considers the completeness and the engineering of the composite material structure dimension definition, and provides a feasible solution with engineering significance for large-scale composite material structure dimension optimization design.
Drawings
In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings needed to be used in the description of the embodiments will be briefly introduced below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without inventive exercise.
FIG. 1 is a flow chart of a composite material layering sequence optimization system provided by the present invention;
FIG. 2 is a schematic diagram of a size definition optimization process provided in the present invention;
FIG. 3 is a schematic diagram of an algorithm flow for converting a constrained optimization problem into an unconstrained optimization problem by using a multiplier penalty function method according to the present invention;
FIG. 4 is a schematic diagram of an algorithm flow for solving an extreme value by using a BFGS quasi-Newton method according to the present invention;
FIG. 5 is a schematic flow chart of the optimized design of layering sequence by using immune genetic algorithm according to the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention clearer, the present invention will be described in further detail with reference to the accompanying drawings, and it is apparent that the described embodiments are only a part of the embodiments of the present invention, not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention optimizes the layering sequence of the composite material by using Aerochelk software. The Aeroceck software is CAE software which is developed completely independently and autonomously, is developed based on Python language, adopts a model-view-controller (MVC) architecture and an SQLite database to perform data management, develops a 3D graphic engine based on VTK, develops a graphic interface based on wxPython, performs a large amount of automatic regression tests and engineering tests, and finally realizes the optimal design of the layering sequence of the composite material.
The embodiments of the present invention will be described in further detail with reference to the drawings attached hereto.
As shown in FIG. 1, the embodiment of the invention discloses a composite material layering sequence optimizing system, which comprises:
the database manufacturing module is used for extracting the original model data according to the created generalized finite element GFEM model, manufacturing the original model data into a database and providing input data for the system;
the composite material strength checking module is used for checking the static strength, the stability and the connection strength of the composite material based on the finite element model;
the size definition optimization module is used for optimizing the size definition of the model according to the user target safety margin and finishing the strength checking iteration;
the size optimization algorithm module is used for converting the constrained optimization problem into an unconstrained optimization problem by adopting a multiplier penalty function method, and solving an extreme value of the unconstrained optimization problem by adopting a BFGS quasi-Newton method;
the automatic layering library optimization module adopts an immune genetic algorithm, combines a composite material structure and a process layering design criterion, performs layering sequence optimization design and automatically generates a layering library;
and the result visualization module checks the intensity checking analysis result through a safety margin cloud picture, a result label or a result table.
Optionally, the safety margin of the embodiment of the present invention refers to a difference between a ratio of failure stress (or strain) to design stress (or design strain) of a material used for a part or a component and 1.
In implementation, a Generalized Finite Element (GFEM) is a conceptual extension of a conventional Finite element Method, and is based on a unit decomposition Method, and by introducing Generalized freedom at a node, the node freedom is interpolated again, so as to improve the approximation accuracy of the Finite element Method or meet special approximation requirements for a specific problem. Based on the deep research of the generalized finite element method on the element shape function construction theory, the complex problems of any internal characteristics (cavities, inclusions, cracks and the like) and external characteristics (concave angles, angular points, edges and the like) are solved on a simple finite element network irrelevant to the region.
Specifically, as shown in fig. 1, in implementation, the calculation is solved according to the GFEM model, and a ply library information table of the ply database creation tool, a load database creation tool f06 file, and geometric information are obtained. The layering and loading information can be checked through a database interactive interface, and unit size definition, material allowable value definition and the like can also be carried out. Finally, after the intensity check is finished, if the safety margin is met, updating the model attribute; and if the safety margin is not met, carrying out size definition optimization again.
The method refines the traditional strength dimension definition process of the aeronautical structure into an automatic optimization solver, fully considers the completeness and the engineering of the composite material structure dimension definition, and provides a feasible solution with engineering significance for large-scale composite material structure dimension optimization design. The invention improves the working efficiency, the specification and the cooperativity of the structural strength check and the size definition of the airplane.
Further, in the above system, the database includes a ply information database, a load information database, and a material information database.
The invention adopts an advanced database management method, effectively manages large-scale engineering data and provides an interactive interface with the existing data of a client.
Further, in the above system, the composite static strength verification includes laminate tensile verification, compressive and shear strain level verification.
Further, in the system, the composite material stability check comprises a one-dimensional unit local instability and pressure loss check, a stiffened plate column buckling check, and a two-dimensional unit compression, shearing and bending instability check.
In practice, composite static strength checks primarily look at the strain levels of the laminate in tension, compression and shear. The most common failure mode is loss of stability when thin-walled/stiffened structures on the wings, empennage, and fuselage of an aircraft are subjected to compressive, shear, torsional, and bending loads. The composite material stability check mainly comprises the steps of one-dimensional unit local instability and pressure loss check, stiffened plate column buckling check, and two-dimensional unit compression, shearing and bending instability check.
Further, as shown in fig. 2, the size definition optimization process is as follows:
1) importing an optimization model;
2) calculating the optimized area according to the specified working condition and the specified constraint;
3) optimizing structural units which do not meet the constraint in the optimized region;
4) if the optimized regions all meet the constraint but do not reach the optimization index (tolerance and iteration times), selecting the structural unit with the maximum constraint value in the optimized regions for optimization until the optimized regions meet the optimization index;
5) exporting the model meeting the optimization index, and submitting the model to a solver Nastran for calculation;
6) updating an optimization model load database;
7) calculating the optimized area according to the specified working condition and the specified constraint, and ending the process if the optimized area meets the optimized indexes; if not, the operation is returned to the step 2 for operation.
Further, as shown in fig. 2, the calculating the optimized region according to the specified working condition and the specified constraint includes: one-dimensional unit working condition circulation and two-dimensional unit working condition circulation;
the one-dimensional unit working condition cycle comprises the following steps:
extracting load information, layering information and allowable value information corresponding to the one-dimensional unit;
performing one-dimensional unit load judgment and rib plate judgment after compression according to one-dimensional unit static strength check, rib plate column buckling, one-dimensional unit local stability and one-dimensional unit pressure loss;
the two-dimensional unit working condition cycle comprises the following steps:
extracting load information, layering information and allowable value information corresponding to the two-dimensional unit;
and judging the two-dimensional unit load according to the static strength check of the two-dimensional unit, the compression-shear combination of the two-dimensional unit and the bending-shear combination of the two-dimensional unit.
In specific implementation, the size definition optimization module optimizes the model size definition according to the user target safety margin, and helps a user to complete the strength check iteration quickly.
Further, as shown in fig. 3, the algorithm flow of the size optimization algorithm module converting the constrained optimization problem into the unconstrained optimization problem by using the multiplier penalty function method is as follows:
1) selecting an initial point X0∈RnInitial Lagrange coefficient (. mu.)i)1∈Rl、(λi)1∈RmInitial penalty factor σ1Greater than 0, tolerance 0 < epsilon-1, contrast ratio
The penalty factor increment coefficient eta is more than 1, and k is equal to 1;
2) with Xk-1For the initial point, the method of BFGS is adopted to solve min psi (X, (. mu.) (μ)i)k，(λi)k，σk) Has an extreme value of Xk；
3) Computing
If beta iskStopping iteration and outputting X when epsilon is less than or equal tokCalculating a minimum value f (X) for an approximate minimum point of the primitive functionk) Otherwise, executing step 4;
4) update the penalty factor if
Let sigmak+1＝ησkElse σk+1＝σkAnd updating the Lagrange coefficient, and calculating according to the following formula:
(μi)k+1＝(μi)k-σk+1hi(Xk)
(λi)k+1＝max{0,(λi)k-gi(Xk)}
5) and k is k +1, and the operation of the step 2 is returned.
The basic idea of the multiplier method adopted by the embodiment of the invention is to start from the Lagrange function of the original problem and add a proper penalty function, thereby converting the original problem into a series of unconstrained optimization sub-problems.
For the general problem:
it is converted to min ψ (x) by augmenting the lagrange multiplier, i.e.:
wherein sigmakFor the penalty factor, (. mu.), (μ)i)k、(λi)kReferred to as lagrange multipliers.
Further, as shown in fig. 4, the size optimization algorithm module adopts a BFGS quasi-newton method to solve the extreme value as follows:
1) given a parameter δ ∈ (0,1), σ ∈ (0,1), and an initial point X0∈RnInitial positive definite matrix B0(taking the identity matrix InOr G (X)0) 0 < epsilon-1, let k be 0;
2) calculate gk＝▽f(Xk) If g | | |kLess than or equal to epsilon, convergence and X outputkCalculating a minimum value f (X) for an approximate minimum point of the primitive functionk) Otherwise, executing step 3;
3) solving a system of linear equations Bkd＝-gkTo obtain a solution dkLet m bekIs the smallest non-negative integer m that satisfies the following inequality:
order to
Xk+1＝Xk+αkdkB is determined by the BFGS correction formulak+1；
4) And k is k +1, and the operation of the step 2 is returned.
Further, in the above systemThe BFGS correction formula
The following were used:
wherein a displacement sk＝xk+1-xkDifference of displacement yk＝gk+1-gk。
The basic idea of the embodiment of the invention for the unconstrained optimization problem Newton's method is to use the iteration point XkThe first derivative (gradient) and the second derivative (Hesee matrix) of the point approximate the quadratic function of the objective function, then the minimum point of the quadratic model is used as a new iteration point, and the process is continuously repeated until the approximate minimum point meeting the precision is obtained.
The quasi-Newton method is to pass through some approximate matrix BkThe Hesee array is replaced, and the problem that the algorithm cannot continue to be used when the Hesee array is singular is avoided. The BFGS rule is based on the approximation matrix B proposed by Broyden, Fletcher, Goldfarb and ShannokThe correction method corrects the same.
Further, as shown in fig. 5, the flow of the layering library automatic optimization module adopting the immune genetic algorithm to perform layering sequence optimization design is as follows:
1) gene coding, namely converting the layering sequence into a chromosome string according to the layering sequence theta i, i is 1, 2, 3.n.n, wherein n is the layering number, the layering angle coding length of each layer is n, integer coding is adopted, the angle corresponding to the layering coding is fixed, and the optimized layering can obtain the corresponding angle after decoding;
specifically, as shown in fig. 5, an initial population is randomly generated by setting basic parameters, a population size, a crossover probability, a mutation probability, a gene string length, a convergence criterion, and a concentration determination threshold, and a gene string is adjusted according to constraints.
The embodiment of the invention combines the coding characteristics of the genetic algorithm, and because the layering angles are selected among fixed numerical values, the whole layering sequence is used as a variable and is converted into a group of chromosome strings, each gene in the chromosome strings represents the layering angle of each layer, the selection is carried out in the fixed numerical values, and the position sequence of the chromosome strings represents the layering sequence from outside to inside.
Therefore, the number of layers is determined according to the layer sequence θ i, i is 1, 2, 3. The stacking sequence is converted into a chromosome string, the stacking angle coding length of each layer is n, and the stacking number of each stacking angle is ensured to be fixed due to genetic crossing operation, so that integer coding is adopted, repeated gene codes are not allowed to select an initial stacking sequence in the chromosome coding of each individual, the angle corresponding to the stacking coding is fixed, and the optimized stacking can obtain the corresponding angle after decoding.
2) Fitness analysis, which adjusts the gene string according to constraints, and by chromosome string adjustment, no invalid chromosomes are generated, so that the objective function value is directly used as the fitness function value, namely:
f＝Nxcr
specifically, as shown in fig. 5, fitness analysis is performed according to the optimization target and the constraint equation, and whether the population converges is determined, if yes, the process is ended, otherwise, the antibody-antigen affinity, the antibody concentration, and the antibody selection probability in step 3 are calculated.
3) Genetic operation, wherein the objective function is antigen, the layering sequence is antibody, the population selection probability is determined according to the promotion and inhibition effect between the antibody and the antigen, and then the selection, crossing and mutation operations are carried out.
In the immune genetic algorithm, the problem to be solved is taken as the antigen of an immune system, the independent variable of the optimized design is taken as an antibody, the objective function of the invention is the antigen, and the layering sequence is the antibody. Determining the population selection probability according to the promotion and inhibition effect between the antibody and the antigen, and then performing selection, crossing and mutation operations.
In the implementation of the invention, in the design of the layering of the composite material laminated plate, the thickness of each layering of the laminated plate must be integral multiple of the thickness of a single-layer material due to the manufacturing manufacturability requirement, the laying angle is usually composed of standard angles of 0 degree, 45 degrees, 90 degrees and the like, and the arrangement sequence of each layering can be changed randomly, so the optimization of the layering sequence of the composite material laminated structure actually belongs to the combined optimization problem of discrete variables, and the traditional optimization method is difficult to solve. The genetic algorithm is a highly parallel, random and self-adaptive search algorithm developed by taking advantage of natural selection and evolution mechanisms in the biology world. Because of its robustness, it is particularly suited to dealing with complex, non-linear problems that are not well solved by traditional search algorithms.
The invention adopts an immune genetic algorithm, considers the structural and technological layering design criteria, automatically carries out layering sequence optimization design, automatically generates a layering library, converts the design variable of the composite material structure into a layering discrete variable in the layering library, greatly reduces the optimization scale, ensures that the optimization of Aerochelk directly associates the layering design of the optimization result with engineering while solving the problem of large-scale optimization design of the composite material structure, fully considers the completeness and the engineering of the composite material structure size definition, and provides a feasible solution with engineering significance for the large-scale optimization design of the composite material structure size.
In conclusion, the composite material layering sequence optimization system applies Aerochel software to solve two fundamental problems of overlarge optimization design scale of composite material structure size, and composite material structure layering and layering sequence design, extracts the traditional strength analysis process, adopts a multiplier-function penalty method to convert a constrained optimization problem into an unconstrained optimization problem, adopts a BFGS quasi-Newton method to solve an extreme value of the unconstrained optimization problem, converts the extreme value into an automatic optimization solver, adopts an immune genetic algorithm, automatically performs layering sequence optimization design by considering the structure and process layering design criteria, automatically generates a layering library, converts design variables of a composite material structure into layering discrete variables in the layering library, greatly reduces the optimization scale, enables the optimization of the Aerochel to solve the composite material structure large-scale optimization design problem and simultaneously directly associates the optimized result layering design with engineering, the completeness and engineering of the composite material structure size definition are fully considered, and a feasible solution with engineering significance is provided for large-scale composite material structure size optimization design.
While preferred embodiments of the present invention have been described, additional variations and modifications in those embodiments may occur to those skilled in the art once they learn of the basic inventive concepts. Therefore, it is intended that the appended claims be interpreted as including preferred embodiments and all such alterations and modifications as fall within the scope of the invention.
It will be apparent to those skilled in the art that various changes and modifications may be made in the present invention without departing from the spirit and scope of the invention. Thus, if such modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include such modifications and variations.
Claims (10)
1. A composite layup order optimization system, comprising:
the database manufacturing module is used for extracting the original model data according to the created generalized finite element GFEM model, manufacturing the original model data into a database and providing input data for the system;
the composite material strength checking module is used for checking the static strength, the stability and the connection strength of the composite material based on the finite element model;
the size definition optimization module is used for optimizing the size definition of the model according to the user target safety margin and finishing the strength checking iteration;
the size optimization algorithm module is used for converting the constrained optimization problem into an unconstrained optimization problem by adopting a multiplier penalty function method, and solving an extreme value of the unconstrained optimization problem by adopting a BFGS quasi-Newton method;
the automatic layering library optimization module adopts an immune genetic algorithm, combines a composite material structure and a process layering design criterion, performs layering sequence optimization design and automatically generates a layering library;
and the result visualization module checks the intensity checking analysis result through a safety margin cloud picture, a result label or a result table.
2. The system of claim 1, wherein the databases comprise a ply information database, a load information database, and a material information database.
3. The system of claim 1, wherein the composite static strength checks comprise laminate tensile checks, compressive and shear strain level checks.
4. The system according to claim 1, wherein the composite stability checks include one-dimensional cell local instability and pressure loss checks, stiffened plate column buckling checks, and two-dimensional cell compression, shear, and bending instability checks.
5. The system of claim 1, wherein the size definition optimization procedure is as follows:
1) importing an optimization model;
2) calculating the optimized area according to the specified working condition and the specified constraint;
3) optimizing structural units which do not meet the constraint in the optimized region;
4) if the optimized regions all meet the constraint but do not reach the optimization index (tolerance and iteration times), selecting the structural unit with the maximum constraint value in the optimized regions for optimization until the optimized regions meet the optimization index;
5) exporting the model meeting the optimization index, and submitting the model to a solver Nastran for calculation;
6) updating an optimization model load database;
7) calculating the optimized area according to the specified working condition and the specified constraint, and ending the process if the optimized area meets the optimized indexes; if not, the operation is returned to the step 2 for operation.
6. The system of claim 5, wherein calculating the optimization region according to the specified conditions and the specified constraints comprises: one-dimensional unit working condition circulation and two-dimensional unit working condition circulation;
the one-dimensional unit working condition cycle comprises the following steps:
extracting load information, layering information and allowable value information corresponding to the one-dimensional unit;
performing one-dimensional unit load judgment and rib plate judgment after compression according to one-dimensional unit static strength check, rib plate column buckling, one-dimensional unit local stability and one-dimensional unit pressure loss;
the two-dimensional unit working condition cycle comprises the following steps:
extracting load information, layering information and allowable value information corresponding to the two-dimensional unit;
and judging the two-dimensional unit load according to the static strength check of the two-dimensional unit, the compression-shear combination of the two-dimensional unit and the bending-shear combination of the two-dimensional unit.
7. The system of claim 1, wherein the algorithm for transforming the constrained optimization problem into the unconstrained optimization problem by the size optimization algorithm module using the multiplier penalty function is as follows:
1) selecting an initial point X0∈RnInitial Lagrange coefficient (. mu.)i)1∈Rl、(λi)1∈RmInitial penalty factor σ1The tolerance is more than 0, the tolerance is more than 0 and less than epsilon, the epsilon is 1, the contrast coefficient theta is epsilon (0,1), the penalty factor growth coefficient eta is more than 1, and k is equal to 1;
2) with Xk-1For the initial point, the method of BFGS is adopted to solve min psi (X, (. mu.) (μ)i)k，(λi)k，σk) Has an extreme value of Xk；
3) Computing
If beta iskStopping iteration and outputting X when epsilon is less than or equal tokCalculating a minimum value f (X) for an approximate minimum point of the primitive functionk) Otherwise, executing step 4;
4) updating the penalty factor if betak≥θβk-1Let σ bek+1＝ησkElse σk+1＝σkAnd updating the Lagrange coefficient, and calculating according to the following formula:
(μi)k+1＝(μi)k-σk+1hi(Xk)
(λi)k+1＝max{0,(λi)k-gi(Xk)}
5) and k is k +1, and the operation of the step 2 is returned.
8. The system of claim 1, wherein the size optimization algorithm module adopts a BFGS quasi-Newton method to solve the extreme value as follows:
1) given a parameter δ ∈ (0,1), σ ∈ (0,1), and an initial point X0∈RnInitial positive definite matrix B0(taking the identity matrix InOr G (X)0) 0 < epsilon-1, let k be 0;
2) calculate gk＝▽f(Xk) If | gk| < epsilon |, convergence, output XkCalculating a minimum value f (X) for an approximate minimum point of the primitive functionk) Otherwise, executing step 3;
3) solving a system of linear equations Bkd＝-gkTo obtain a solution dkLet m bekIs the smallest non-negative integer m that satisfies the following inequality:
order toXk+1＝Xk+αkdkB is determined by the BFGS correction formulak+1；
4) And k is k +1, and the operation of the step 2 is returned.
9. The system of claim 8, wherein the BFGS correction formula
The following were used:
wherein a displacement sk＝xk+1-xkDifference of displacement yk＝gk+1-gk。
10. The system of claim 1, wherein the automatic layering library optimization module adopts an immune genetic algorithm to perform a layering sequence optimization design process as follows:
1) gene coding according to the layering order thetaiN, wherein n is the number of layers, the layer sequence is converted into a chromosome string, the layer angle coding length of each layer is n, integer coding is adopted, the angle corresponding to the layer coding is fixed, and the optimized layer can obtain the corresponding angle after being decoded;
2) fitness analysis, which adjusts the gene string according to constraints, and by chromosome string adjustment, no invalid chromosomes are generated, so that the objective function value is directly used as the fitness function value, namely:
f＝Nxcr
3) genetic operation, wherein the objective function is antigen, the layering sequence is antibody, the population selection probability is determined according to the promotion and inhibition effect between the antibody and the antigen, and then the selection, crossing and mutation operations are carried out.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810699919.2A CN110728081B (en) | 2018-06-29 | 2018-06-29 | Composite material layering sequence optimizing system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810699919.2A CN110728081B (en) | 2018-06-29 | 2018-06-29 | Composite material layering sequence optimizing system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110728081A true CN110728081A (en) | 2020-01-24 |
| CN110728081B CN110728081B (en) | 2023-08-29 |
Family
ID=69216763
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810699919.2A Active CN110728081B (en) | 2018-06-29 | 2018-06-29 | Composite material layering sequence optimizing system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110728081B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111985017A (en) * | 2020-03-03 | 2020-11-24 | 广东三维家信息科技有限公司 | Method and device for adapting product to wall opening and electronic equipment |
| CN114722509A (en) * | 2022-06-09 | 2022-07-08 | 华中科技大学 | Conical reinforced cabin layering sequence optimization method based on fiber continuous model |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140365180A1 (en) * | 2013-06-05 | 2014-12-11 | Carnegie Mellon University | Optimal selection of building components using sequential design via statistical based surrogate models |
| CN106156449A (en) * | 2016-08-31 | 2016-11-23 | 中航沈飞民用飞机有限责任公司 | A kind of composite wing wallboard Optimization Design |
| CN106339523A (en) * | 2016-06-24 | 2017-01-18 | 北京玻钢院复合材料有限公司 | Design method of large-torque composite transmission shaft structure |
-
2018
- 2018-06-29 CN CN201810699919.2A patent/CN110728081B/en active Active
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20140365180A1 (en) * | 2013-06-05 | 2014-12-11 | Carnegie Mellon University | Optimal selection of building components using sequential design via statistical based surrogate models |
| CN106339523A (en) * | 2016-06-24 | 2017-01-18 | 北京玻钢院复合材料有限公司 | Design method of large-torque composite transmission shaft structure |
| CN106156449A (en) * | 2016-08-31 | 2016-11-23 | 中航沈飞民用飞机有限责任公司 | A kind of composite wing wallboard Optimization Design |
Non-Patent Citations (1)
| Title |
|---|
| 赵艳雄;何景武;胡哲;: "基于GEMGA的复合材料层合板结构优化设计研究", 强度与环境, no. 01 * |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN111985017A (en) * | 2020-03-03 | 2020-11-24 | 广东三维家信息科技有限公司 | Method and device for adapting product to wall opening and electronic equipment |
| CN111985017B (en) * | 2020-03-03 | 2024-04-09 | 广东三维家信息科技有限公司 | Method and device for adapting product to wall hole and electronic equipment |
| CN114722509A (en) * | 2022-06-09 | 2022-07-08 | 华中科技大学 | Conical reinforced cabin layering sequence optimization method based on fiber continuous model |
| CN114722509B (en) * | 2022-06-09 | 2022-09-09 | 华中科技大学 | Conical reinforced cabin layering sequence optimization method based on fiber continuous model |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110728081B (en) | 2023-08-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Dababneh et al. | Application of an efficient gradient-based optimization strategy for aircraft wing structures | |
| CN110807154A (en) | Recommendation method and system based on hybrid deep learning model | |
| Hao et al. | Adaptive gradient-enhanced kriging model for variable-stiffness composite panels using Isogeometric analysis | |
| Wang et al. | Lightweight design of variable-angle filament-wound cylinders combining Kriging-based metamodels with particle swarm optimization | |
| Passos et al. | Optimal curved fibre orientations of a composite panel with cutout for improved buckling load using the Efficient Global Optimization algorithm | |
| Dillinger et al. | Static aeroelastic stiffness optimization of a forward swept composite wing with CFD-corrected aero loads | |
| US10896270B2 (en) | Method for solving multi-fidelity optimization problems | |
| CN110728081B (en) | Composite material layering sequence optimizing system | |
| Pereira et al. | Deep multiobjective design optimization of CFRP isogrid tubes using lichtenberg algorithm | |
| Macquart et al. | Aeroelastic design of blended composite structures using lamination parameters | |
| Irisarri et al. | A general optimization strategy for composite sandwich structures | |
| Zhao et al. | Buckling analysis and optimization of stiffened variable angle tow laminates with a cutout considering manufacturing constraints | |
| El Said | Predicting the non-linear response of composite materials using deep recurrent convolutional neural networks | |
| Kim et al. | A novel adaptive mesh refinement scheme for the simulation of phase‐field fracture using trimmed hexahedral meshes | |
| CN103353916A (en) | Method for post-processing engineering-based composite material laminated board ply after optimization | |
| De Gaspari | Multiobjective optimization for the aero-structural design of adaptive compliant wing devices | |
| Hui et al. | A data-driven CUF-based beam model based on the tree-search algorithm | |
| Barkanov et al. | Weight optimal design of lateral wing upper covers made of composite materials | |
| Wang et al. | Data-driven inverse method with uncertainties for path parameters of variable stiffness composite laminates | |
| Wu et al. | An integrated approach to shape and laminate stacking sequence optimization of free-form FRP shells | |
| Wang et al. | Stiffness and strength topology optimization for bi‐disc systems based on dual sequential quadratic programming | |
| Zein et al. | A repair operator for the preliminary design of a composite structure using genetic algorithms | |
| CN116303386A (en) | Intelligent interpolation method and system for missing data based on relational graph | |
| CN110659525B (en) | Optimized design method for composite structure | |
| Chauhan et al. | Wing shape optimization using FFD and twist parameterization |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |