Embedded optical fiber composite material optimization method based on adaptive genetic algorithm
Through the two-level optimization strategy of adaptive genetic algorithm, the fiber sensor layout and composite material laying design are optimized, which solves the problems of high computational cost and low optimization efficiency of embedded fiber composite materials, and achieves the coordinated optimization of sensing coverage and structural performance.
Patent Information
- Application Number
- CN202510665794.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-05-22
- Publication Date
- 2025-07-04
AI Technical Summary
When optimizing embedded optical fiber composite materials, the prior art has high computing cost and low optimization efficiency, and cannot effectively take into account both sensing and structural performance.
A two-level optimization strategy based on adaptive genetic algorithm is adopted, firstly the layout of the fiber sensor is optimized to meet the sensing coverage requirements, and then the laying design of composite laminates is optimized, combining adaptive crossover and mutant operators to improve optimization efficiency.
It effectively reduces the calculation cost, improves optimization efficiency, realizes coordinated optimization of sensing coverage, structural lightweighting and mechanical performance, and solves the contradiction between sensor system complexity and structural safety.
Smart Images

Figure CN120257837A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of fiber-reinforced composite materials, and particularly to an optimization method for embedded optical fiber composite materials based on an adaptive genetic algorithm. Background Technique
[0002] Fiber-reinforced composite materials are a research hotspot in the fields of materials science and engineering and have been widely used in the aerospace field. During the in-service period of composite materials, certain internal damages may occur, which are difficult to directly observe from the surface. Therefore, it is very important to conduct real-time health monitoring of the internal structure of composite materials and accurately grasp the changes in real-time stress, strain and other information.
[0003] The structure of embedded optical fiber composite materials integrates optical fiber sensors with composite materials, which can not only exert the performance of the original composite materials, but also monitor the deformation and damage inside the structure in real time. The optical fiber sensors are buried in the composite materials to form an integrated structure, and may face the compatibility problem of the integrated structure of multi-component materials in a complex working environment. After the optical fiber sensors are buried in the composite materials, the high-temperature and high-pressure curing during the manufacturing process will affect the performance of the optical fiber sensors, which is one aspect of compatibility. On the other hand, resin-rich areas will be formed around the optical fibers and fibers after burial, which will affect the mechanical properties of the structure.
[0004] In the process of designing intelligent composite materials, it is necessary to minimize the negative impact of optical fibers on the performance of composite materials to the greatest extent, and at the same time, the health monitoring function of optical fibers should also be taken into account. At present, the optical fiber layout optimization method mainly considers the sensing range and accuracy of optical fibers and the influence of internal stress in the optical fiber structure, and usually does not consider the influence of optical fiber arrangement on the performance of the overall structure; the composite material optimization method usually considers structural performance parameters such as stiffness and strength.
[0005] For embedded optical fiber composite materials, in order to meet the sensing requirements, a large number of optical fiber sensors need to be arranged, thus forming a complex sensing system; at the same time, the buried optical fiber sensors may affect the continuity of the composite materials, resulting in a reduction in mechanical properties. Therefore, embedded optical fiber composite materials must consider both mechanical properties and sensing properties at the same time. There is a high-dimensional design space for the optimization of embedded optical fiber composite materials, and the design variables include parameters such as composite material laminates (such as ply angles, ply sequences, ply thicknesses) and sensor layouts (such as the number and positions of sensors). A large number of design parameters lead to huge computational costs and low optimization efficiency. Summary of the Invention
[0006] The purpose of the present application is to provide an optimization method for embedded optical fiber composite materials based on an adaptive genetic algorithm, which can effectively reduce the computational cost and improve the optimization efficiency.
[0007] To achieve the above purpose, the present application provides the following solutions:
[0008] This application provides an optimization method for embedded fiber optic composites based on an adaptive genetic algorithm. The embedded fiber optic composites include fiber optic sensors and composite laminates. The optimization method for embedded fiber optic composites based on an adaptive genetic algorithm includes:
[0009] Select the layout parameters of the fiber optic sensors as design variables, with the goal of minimizing the number of fiber optic sensors that meet the constraint condition of covering the entire area to be monitored for sensing, and establish a sensor layout optimization model;
[0010] Use the adaptive genetic algorithm to solve the sensor layout optimization model to obtain the optimal layout of the fiber optic sensors; the adaptive genetic algorithm uses an adaptive crossover operator and an adaptive mutation operator; the adaptive crossover operator and the adaptive mutation operator adaptively determine the crossover probability and mutation probability based on the relationship between the population average fitness and the individual fitness and based on the number of iterations;
[0011] Based on the optimal layout of the fiber optic sensors, select the ply parameters of the composite laminate as design variables, with the structural non - failure and sensing covering the entire area to be monitored as constraint conditions, and minimize the total structural weight and displacement as the optimization goal, and establish a composite laminate optimization model;
[0012] Use the adaptive genetic algorithm to solve the composite laminate optimization model to obtain the optimal ply design scheme of the composite laminate;
[0013] Optimize the embedded fiber optic composites according to the optimal ply design scheme of the composite laminate and the optimal layout of the fiber optic sensors.
[0014] According to the specific embodiments provided by this application, this application has the following technical effects:
[0015] This application provides an optimization method for embedded fiber optic composites based on an adaptive genetic algorithm. For the optimal design of fiber optic sensors and composite laminates in embedded fiber optic composites, a two - stage optimization strategy is adopted: in the first - stage optimization, select the layout parameters of the fiber optic sensors as design variables, consider the constraint condition of covering the entire area to be monitored for sensing, and optimize the layout of the fiber optic sensors; in the second - stage optimization, under the constraint of the optimal layout of the fiber optic sensors, optimize the ply design of the composite laminate, avoid the problem of computational explosion caused by global search, reduce the amount of calculation, thereby effectively reducing the computational cost and improving the optimization efficiency. BRIEF DESCRIPTION OF THE DRAWINGS
[0016] To more clearly illustrate the technical solutions in the embodiments of the present application or related technologies, the following will briefly introduce the accompanying drawings required in the embodiments. Obviously, the accompanying drawings in the following description are only some embodiments of the present application. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.
[0017] Figure 1 Schematic flow chart of an optimization method for embedded optical fiber composites based on an adaptive genetic algorithm provided by an embodiment of the present application;
[0018] Figure 2 Schematic diagram of geometric shapes and boundary conditions provided by a specific embodiment of the present application;
[0019] Figure 3 Optimal design displacement nephogram provided by a specific embodiment of the present application;
[0020] Figure 4 Optimal design failure factor nephogram provided by a specific embodiment of the present application;
[0021] Figure 5 Schematic flow chart in more detail of an optimization method for embedded optical fiber composites based on an adaptive genetic algorithm provided by an embodiment of the present application. Specific embodiments
[0022] The following will clearly and completely describe the technical solutions in the embodiments of the present application with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, rather than all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.
[0023] To make the above objects, features, and advantages of the present application more obvious and understandable, the present application will be further described in detail below with reference to the accompanying drawings and specific embodiments.
[0024] In an exemplary embodiment, as Figure 1 shown, an optimization method for embedded optical fiber composites based on an adaptive genetic algorithm is provided. The embedded optical fiber composites include optical fiber sensors and composite laminates. The method includes the following steps 101 to 105. Among them:
[0025] Step 101: Select the layout parameters of the optical fiber sensors as design variables, and establish a sensor layout optimization model with the goal of minimizing the number of optical fiber sensors that meet the constraint condition of covering the entire area to be monitored.
[0026] Step 102: Solve the sensor layout optimization model using the adaptive genetic algorithm to obtain the optimal layout of the fiber optic sensors; the adaptive genetic algorithm uses an adaptive crossover operator and an adaptive mutation operator; the adaptive crossover operator and the adaptive mutation operator adaptively determine the crossover probability and the mutation probability based on the relationship between the population average fitness and the individual fitness and based on the number of iterations.
[0027] Step 103: Based on the optimal layout of the fiber optic sensors, select the ply parameters of the composite laminate as the design variables, with the constraints that the structure does not fail and the sensing covers the entire area to be monitored, and minimize the total weight and displacement of the structure as the optimization objectives to establish an optimization model for the composite laminate.
[0028] Step 104: Solve the optimization model of the composite laminate using the adaptive genetic algorithm to obtain the optimal ply design scheme of the composite laminate.
[0029] Step 105: Optimize the embedded fiber composite material according to the optimal ply design scheme of the composite laminate and the optimal layout of the fiber optic sensors.
[0030] In the multi-level optimization strategy, the global optimization problem is systematically decomposed into several sub-problems, and then these problems are gradually solved. Implementing the above steps 101 to 105, the present application proposes a two-level optimization algorithm. In the first-level optimization (steps 101 to 102), the layout of the fiber optic sensors (such as distributed fiber optic sensors) is selected as the design variable, and the structural strength and sensing performance are considered as constraints; in the second-level optimization (steps 103 to 104), the structural strength, sensing performance, and the number of sensors are considered as constraints, and the two aspects of the optimal design of the sensor layout and the optimal design of the composite material structure are organically combined, while effectively reducing the complexity of the finite element calculation and improving the efficiency of the algorithm.
[0031] In another exemplary embodiment of the present application, in the fiber optic sensing network, only when all the points to be monitored are fully monitored and covered, that is, the coverage rate reaches 100%, can the efficient operation and accurate performance of the sensing network be ensured. The specific calculation method of the sensor monitoring coverage rate is as follows: when the point to be monitored is within a distance R1 from the fiber optic sensor, the data obtained by the interpolation method has high accuracy, and when the point to be monitored is outside a distance R2 from the fiber optic sensor, the accuracy of the interpolated data is low. When the point to be monitored is between R1 and R2, its monitoring coverage rate follows a certain probability distribution. For simplicity of calculation, it is considered that the probability of the point to be monitored being monitored and covered is inversely proportional to the distance between this point and the fiber optic sensor. Therefore, the process of determining the constraint condition for the sensing to cover the entire area to be monitored includes the following steps 201 to 204:
[0032] Step 201: Select multiple points to be monitored on the composite material; the points to be monitored are ideal location points for embedding fiber optic sensors.
[0033] Step 202: Use the formula to calculate the monitoring probability of each point to be monitored by one fiber optic sensor.
[0034] In the formula, (x i , y i ) is the coordinate of the i-th fiber optic sensor, (x k , y k ) is the coordinate of the k-th point to be monitored, P(x k , y k , x i , y i ) is the monitoring probability of the k-th point to be monitored by the i-th fiber optic sensor, R is the distance between the node where the i-th fiber optic sensor is located and the k-th point to be monitored, R1 is the first distance threshold, and R2 is the second distance threshold.
[0035] Step 203: According to the monitoring probability of each point to be monitored by one fiber optic sensor, use the formula to determine the monitoring probability of each point to be monitored by a sensing system composed of multiple fiber optic sensors.
[0036] In the formula, P(x k , y k ) is the monitoring probability of the k-th point to be monitored by a sensing system composed of multiple fiber optic sensors, and m' is the number of fiber optic sensors in the sensing system.
[0037] Step 204: According to the monitoring probability of each point to be monitored by a sensing system composed of multiple fiber optic sensors, take the average value of the monitoring probabilities of all points to be monitored by the sensing system composed of multiple fiber optic sensors as the global monitoring probability, and aim for the coverage rate of all points to be monitored to reach 100%, and obtain the constraint condition for the sensing to cover the entire area to be monitored as In the formula, POD is the monitoring coverage rate of the fiber optic sensor, and n is the number of fiber optic sensors.
[0038] In another exemplary embodiment of the present application, the sensor layout optimization model is:
[0039]
[0040] In the formula, S j is a binary variable indicating whether the j-th candidate point is embedded with a fiber optic sensor. When the j-th candidate point is embedded with a fiber optic sensor, S j takes the value of 1. When the j-th candidate point is not embedded with a fiber optic sensor, S jThe value is 0; J is the number of candidate points; Y is a vector composed of the layout parameters of the fiber optic sensors, and y i is the layout parameter of the i-th fiber optic sensor, n is the number of fiber optic sensors, and POD is the monitoring coverage rate of the fiber optic sensors.
[0041] The number of points to be monitored is greater than the number of candidate points. The points to be monitored are the position points where fiber optic sensors are ideally buried, while the candidate points are the candidate position points for burying fiber optic sensors selected in this application.
[0042] In another exemplary embodiment of this application, the detailed process of the above step 102 is as follows:
[0043] Step 301: Initialize the position parameters of the fiber optic sensors for the first-level optimization, generate an initial population containing multiple layout schemes, and calculate the monitoring probabilities of each layout.
[0044] The binary coding method is used to describe the layout positions of the distributed fiber optic sensors. Several candidate points are selected in the structure to be monitored, and the optimal design is based on whether to bury fiber optic sensors at the candidate points. The total number of distributed fiber optic sensors used in the system is:
[0045]
[0046] In the formula, the variable J is introduced to represent the total number of candidate points on the structure, and j is used to identify the specific number of each candidate point. In addition, a binary variable S j is introduced, and its value is 0 or 1, aiming to reflect whether a fiber optic sensor is buried at a specific candidate point j. Specifically, if a fiber optic sensor is buried at candidate point j, the value of this variable is assigned 1; otherwise, if no fiber optic sensor is buried at this point, the value is assigned 0.
[0047] Randomly generate an initial population of multiple layout schemes, and adopt the elite strategy to select the best-performing individual as the elite, which serves as the initial population of the genetic algorithm.
[0048] The purpose of the first-level optimization of the embedded fiber composite material is to find the minimum number of fiber optic sensors required to meet the constraint conditions. Therefore, the objective function of the first-level optimization of the embedded fiber composite material is:
[0049]
[0050] The penalty infeasible solution is used to transform the constrained problem into an unconstrained problem. The core idea of this method is that for an individual for which no corresponding feasible solution can be found in the solution space, a penalty function is introduced when calculating its fitness to reduce its fitness value, thereby reducing the possibility of this individual being inherited into the next generation population. For the given constraint conditions, a multiplicative form of the evaluation function is selected:
[0051] E v = F v × P v ;
[0052] Wherein, E v is the evaluation function, F v is the fitness, and P v is the penalty function.
[0053] The evaluation function is a global evaluation index that synthesizes the fitness and the penalty function. In the genetic algorithm, only the evaluation function is used to quantify the performance of candidate solutions, and at the same time, the constrained problem is transformed into an unconstrained problem.
[0054] In the fiber optic sensing network, only when all the points to be monitored are fully covered, that is, the coverage rate reaches 100%, can the efficient operation and accurate performance of the sensing network be ensured. Therefore, the first-level optimization of the embedded fiber optic composite material considers the sensing performance as a constraint condition:
[0055] POD = 1.
[0056] Step 302: Optimize the number of sensors using the adaptive crossover and mutation operators, and iteratively solve for the minimum number of sensor configurations that satisfy the constraint conditions.
[0057] The crossover and mutation operators are respectively used to optimize the number of fiber optic sensors.
[0058] Among them, the crossover operation involves two paired chromosomes, and exchanges some of their genes in a certain way to generate two new individuals. The single-point crossover first randomly determines the position of a crossover point, and then exchanges the gene segments after the crossover point of the two individuals participating in the crossover to form two new individuals. The crossover operation is controlled by the crossover probability, which can improve the diversity of the offspring population and effectively prevent the optimization process from prematurely falling into a local optimal solution.
[0059] Among them, the mutation operation involves selecting specific genes in the individual coding string and replacing them to generate new individuals. For individuals using binary coding, the mutation operation means taking the inverse operation on the coding at the mutation point.
[0060] Due to problems such as premature convergence, insufficient global optimization ability, and slow population evolution speed in the basic genetic algorithm, this application designs an adaptive genetic algorithm.
[0061] The adaptive genetic algorithm optimizes the population through extensive search at the initial stage to avoid premature convergence. When approaching the optimal solution in the later stage, the algorithm turns to local fine search and flexibly adjusts the evolutionary strategy. Since the crossover and mutation operators have a crucial impact on the convergence ability of the algorithm, the key to improving the algorithm lies in combining the crossover and mutation operators with the population fitness and the algorithm to construct a new adjustment strategy. In the region close to the optimal solution, the crossover and mutation operations should be appropriately reduced.
[0062] In the crossover operation, first compare the larger fitness value f' among the two individuals participating in the crossover with the average fitness f of the parent population. avg When f' is greater than f, avg it means that this numerical solution is relatively close to the optimal solution. To prevent deviation from the optimal solution due to the crossover operation, the crossover probability needs to be appropriately reduced. Conversely, if the distance between this numerical solution and the optimal solution is still far, to prevent the algorithm from falling into premature convergence, the crossover probability of these two individuals is determined according to the current iteration number; in the mutation operation, since the mutation operator acts on the genes of a single individual, only the individual fitness f needs to be compared with f avg to determine the mutation probability of this individual. In the early stage of iteration, the main goal of the algorithm is to quickly approach the range where the optimal solution is located, and the crossover operation dominates. In the later stage, as the fitness of individuals gradually approaches the optimal solution, the importance of the mutation operation gradually emerges. Based on the above analysis, this application designs an adaptive crossover operator and an adaptive mutation operator to improve the accuracy and efficiency of the genetic algorithm in the optimization process:
[0063]
[0064] In the formula, P c is the crossover probability, f' is the larger fitness value among the two individuals participating in the crossover, f avg is the average fitness of the parent generation, f max is the maximum fitness value of the parent generation, iter is the current iteration number, and iter max is the maximum iteration number. P m is the mutation probability, and the fitness of the offspring individual participating in the mutation operation is f. P c1 , P c2 , P c3 , P m1 , P m2 , P m3 are all adjustable parameters. Exemplarily, to ensure that the operator values are within a reasonable range, select P c1 = 0.9, P c2 = 0.6, P c3 = 0.2, P m1 = 0.05, P m2 = 0.01, P m3= 0.01.
[0065] Step 303: Conduct convergence discrimination based on the relative error of the objective function values in two adjacent iterations. If the termination condition is satisfied, output the optimal sensor layout; otherwise, return to the population update process.
[0066] In another exemplary embodiment of the present application, the optimization model of the composite laminate is:
[0067]
[0068] In the formula, X is a vector composed of the ply parameters of the composite laminate, and x l is the parameter of the l-th ply in the composite laminate, ω1 and ω2 are both weight functions, is the dimensionless weight of the embedded optical fiber composite material, is the dimensionless displacement of the embedded optical fiber composite material, I F is the failure factor, POD is the monitoring coverage rate of the fiber optic sensor, n is the number of fiber optic sensors, n min is the minimum number in the optimal layout of the fiber optic sensors, m is the weight of the embedded optical fiber composite material, u is the displacement of the embedded optical fiber composite material, m0 is the weight of the embedded optical fiber composite material in the initial design, and u0 is the displacement of the embedded optical fiber composite material in the initial design.
[0069] In another exemplary embodiment of the present application, the determination process of the failure factor includes the following steps 401 to 403:
[0070] Step 401: Considering the influence of the fiber optic sensor on the mechanical properties of the structure, construct a finite element model of the composite laminate.
[0071] Step 402: Solve the finite element model of the composite laminate to obtain the mechanical response of the composite material.
[0072] Step 403: Based on the mechanical response of the composite material, and using the formula calculate the failure factor according to the Tsai-Wu failure criterion; in the formula, F 11 , F 22 , F 66 , F1, F2, F 12 are all strength parameters, and F 11 , F 22 , F 66 , F1, F2, F 12 are measured by experiments and generally can also be directly calculated from X t , X c , Y t , Y c , S. σ 11The stress of the composite material in the 1 direction, σ 22 The stress of the composite material in the 2 direction, τ 12 The shear stress of the composite material.
[0073]
[0074] Among them, X t Is the axial or longitudinal tensile ultimate stress, X c Is the axial or longitudinal compressive ultimate stress, Y t Is the transverse tensile ultimate stress, Y c Is the transverse compressive ultimate stress, and S is the shear ultimate stress.
[0075] The key performance parameters of weight, displacement and failure factor can be extracted through the mechanical response of the composite material for optimizing the target evaluation.
[0076] In another exemplary embodiment of the present application, the laying parameters of the composite laminate include: laying angle, laying sequence and laying thickness.
[0077] The laying angle of the composite laminate is described by integer coding. The integer coding corresponds to the laying from the inside to the outside from left to right to intuitively reflect the laying structure. For example: [1,1,2,2,3,3,4,0] → [0,0,45,45, - 45,-45,90], where 0 represents no laying of the corresponding layer, and the thickness of the composite material can be optimized. 1, 2, 3, 4 correspond to laying angles of 0°, 45°, - 45°, 90° respectively.
[0078] In another exemplary embodiment of the present application, the above step 104 is replaced by the following steps 501 to 503:
[0079] Step 501: Establish a weighted fitness function for multi-objective optimization. Combine the structural strength and sensing performance constraints, and use a dynamic penalty function to handle infeasible solutions to generate the fitness scores of population individuals.
[0080] The fitness function is the same as the objective function. The process of calculating the fitness score in step 501 is the same as the process of calculating the fitness value in step 301.
[0081] In step 501, the second-level optimization of the composite material should consider the weight and maximum displacement of the structure. The optimization objective function is:
[0082]
[0083] Each weight parameter should be selected for the optimization variable. The dimensionless weight of the embedded optical fiber composite material And the dimensionless displacement of the embedded optical fiber composite material Are defined as:
[0084]
[0085] Among them, m0 and u0 are calculated according to the initial design.
[0086] In step 501, the effective and safe use of the composite material is crucial for the structural design. Therefore, the second-level optimization of the embedded optical fiber composite material considers the structural strength, sensing performance, and the number of distributed optical fiber sensors as constraints:
[0087]
[0088] Step 502: Adopt a dynamic probability adjustment strategy linked to the number of iterations and individual fitness, and perform the crossover and mutation operations of the genetic algorithm to improve the global search efficiency of the composite material layup parameters.
[0089] The process of performing the genetic algorithm in step 502 is the same as that in step 302.
[0090] Step 503: When the stability of the solution sets for multiple consecutive generations reaches the preset threshold, terminate the iteration and output the optimal layup design scheme.
[0091] Stability means that the change range of the optimal fitness for several consecutive generations is less than the threshold.
[0092] In another exemplary embodiment of the present application, the two-level optimization results are integrated to output a global solution including the minimum number of sensors, the optimal layup sequence, and performance indicators, and a high-precision finite element model is reconstructed based on the optimal parameters to verify the structural safety and monitoring coverage rate.
[0093] The following uses a specific embodiment to illustrate the effectiveness and practicality of the method of the present application.
[0094] As Figure 2 shown, a rectangular design domain with a length of 200 mm and a width of 200 mm is shown in the figure. The four sides are fixed and are subjected to a uniform distributed pressure with a magnitude of 0.1 MPa. Optimize the layout of the optical fiber and the layup parameters of the composite material. The optimization objectives are to minimize the total structural weight and the maximum displacement, and the constraints are that the structure does not fail and the sensing system covers the entire panel.
[0095] The geometry and boundary conditions of the wing panel are as Figure 2 shown. The structure must ensure that it can withstand a pressure load of 0.1 MPa without material failure (the Tsai-Wu failure factor must be less than 1.0), and at the same time, the sensing system must cover the entire panel (the monitoring coverage probability must be equal to 1.0).
[0096] Since the fitness function considered in this embodiment needs to be dimensionless, and regardless of the laminate ply form, the thickness of each single layer is the same, and the weight of the fiber optic sensor is much smaller than that of the laminate, the description of the dimensionless structural weight can be replaced by the number of plies:
[0097] The first-level optimization aims to obtain the minimum number of fiber optic sensors that meet the POD requirements. 19 candidate points are evenly distributed in the x direction, with a spacing of 10 mm between each point. The population size is set to 100, the mutation probability is set to 0.05, and the crossover probability is set to 0.9. After 100 iterations of the optimization design, it is determined that at least seven sensors are required to achieve a POD of 1.
[0098] The second-level optimization aims to obtain the optimal solution that minimizes the objective function. The laying angles of the composite material single-layer plates are selected from {0°, 45°, -45°, 90°}, the population size is set to 100, the mutation probability is set to 0.05, and the crossover probability is set to 0.9. For the fitness weight parameters, take ω1 = ω2 = 0.5. After 100 iterations of the optimization design, the best fiber optic sensor layout is [1001001001001001001], and the best composite material laminate arrangement is [0°, 0°, 0°, 0°, 0°]s.
[0099] A model is established according to the optimal solutions of the above components, and the displacement nephogram and Tsai-Wu failure factor nephogram of the fiber optic intelligent composite wing are obtained by finite element simulation analysis as Figure 3 and Figure 4 shown. From Figure 3 and Figure 4 it can be seen that for the optimized fiber optic intelligent composite wing, the number of plies n = 5, the maximum displacement u = 2.19 mm, and the maximum Tsai-Wu failure factor is 0.26.
[0100] Calculations show that after the fiber optic intelligent composite wing is optimized, its monitoring coverage rate is guaranteed to be 100%, the maximum displacement along the load direction is 4.60 mm, which is 52.4% less than before optimization, that is, the bending stiffness of the structure is improved; the maximum Tsai-Wu failure factor of the structure is 0.26, which is 18.8% less than before optimization, that is, the strength of the structure is improved and the structural material is safer. The differences between the optimal solution and the initial design are compared. As shown in Table 1, compared with the initial design, the optimal solution reduces the maximum displacement by 52.4% and the maximum failure factor by 18.8% at the cost of a 25% increase in weight, and the increase in weight is less than the improvement in performance. The above results show that the optimization strategy adopted in this application is effective and practical.
[0101] Table 1 Comparison between the optimal design and the initial design
[0102] Weight Maximum displacement Maximum failure factor Objective function value Initial design 136g 4.60 mm 0.32 1 Optimal solution 170g 2.19 mm 0.26 0.863
[0103] Figure 5 This is a more detailed flowchart of the method of this application, including the following S1 to S10.
[0104] S1: Construct a monitoring model of a distributed fiber optic sensor in a composite material, calculate the coverage area and monitoring probability of the sensor according to the preset candidate point layout, and achieve the coverage determination of structural health monitoring.
[0105] S2: Initialize the position parameters of the fiber optic sensor optimized at the first level, generate an initial population containing multiple layout schemes, and calculate the fitness values of each layout.
[0106] S3: Based on the sensitivity analysis results of the design variables with respect to the objective function, adopt an adaptive crossover and mutation operator to optimize the number of sensors, and iteratively solve for the minimum sensor configuration that satisfies the constraint conditions.
[0107] S4: Perform convergence discrimination according to the relative error of the objective function values in two adjacent iterations. If the termination condition is satisfied, output the optimal sensor layout; otherwise, return to the population update process.
[0108] S5: Define the set of ply parameters of the composite laminate optimized at the second level, including ply angles, sequence, and thickness, and construct a finite element calculation model integrating the fiber optic sensing network.
[0109] S6: Calculate the mechanical response of the composite material structure through a finite element solver, and extract the key performance parameters such as weight, displacement, and failure factor for optimizing the objective evaluation.
[0110] S7: Establish a weighted fitness function for multi-objective optimization, combine the constraints of structural strength and sensing performance, and handle infeasible solutions through a dynamic penalty function to generate the fitness scores of population individuals.
[0111] S8: Adopt a dynamic probability adjustment strategy linked to the number of iterations and individual fitness, and perform the crossover and mutation operations of the genetic algorithm to improve the global search efficiency of the ply parameters of the composite material.
[0112] S9: When the stability of the solution sets in multiple consecutive generations reaches the preset threshold, terminate the iteration and output the optimal ply design scheme.
[0113] S10: Integrate the optimization results of the two levels, output the global solution including the minimum number of sensors, the optimal ply sequence, and performance indicators, and reconstruct a high-precision finite element model based on the optimal parameters to verify the structural safety and monitoring coverage.
[0114] The beneficial effects of this application are:
[0115] (1) Break through the limitation of the separate optimization of sensing performance and structural performance in traditional methods. Through a two-level optimization strategy, organically integrate the fiber layout and composite material ply design to achieve the global collaborative optimization of multiple objectives such as sensing coverage, structural lightweight, stiffness, and strength, and solve the contradiction between the complexity of the sensing system and the structural safety in embedded fiber composites.
[0116] (2) Adopt a multi-level decomposition strategy to decompose the high-dimensional optimization problem (involving multiple variables such as ply parameters and sensor positions) into two-level sub-problems, reducing the finite element calculation amount. The first-level optimization fixes the ply form to quickly determine the minimum number of sensors, and the second-level optimization conducts structural design under sensor constraints to avoid the computational explosion problem caused by global search.
[0117] (3) Adopt an adaptive genetic algorithm, consider the relationship between the average fitness of the population and the individual fitness, and the connection between the number of iterations and the genetic operator, which has a significant improvement in the optimization ability and solves the defects such as premature convergence easily occurring in the optimization process of the traditional genetic algorithm.
[0118] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity of description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.
[0119] Specific examples are used in this article to elaborate on the principles and implementation manners of this application. The description of the above embodiments is only used to help understand the method and its core idea of this application; at the same time, for those of ordinary skill in the art, according to the idea of this application, there will be changes in the specific implementation manners and application scopes. In summary, the content of this specification should not be construed as a limitation to this application.
Claims
1. An optimization method for embedded optical fiber composites based on an adaptive genetic algorithm, characterized in that, The embedded optical fiber composite material includes an optical fiber sensor and a composite laminate. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm includes: Select the layout parameters of the optical fiber sensor as design variables, and establish a sensor layout optimization model with the goal of minimizing the number of optical fiber sensors that meet the constraint condition of sensing the entire area to be monitored. Solve the sensor layout optimization model using the adaptive genetic algorithm to obtain the optimal layout of the optical fiber sensor; the adaptive genetic algorithm uses an adaptive crossover operator and an adaptive mutation operator; the adaptive crossover operator and the adaptive mutation operator adaptively determine the crossover probability and mutation probability based on the relationship between the population average fitness and the individual fitness and based on the number of iterations. Based on the optimal layout of the optical fiber sensor, select the ply parameters of the composite laminate as design variables, and establish a composite laminate optimization model with the goal of minimizing the total structural weight and displacement under the constraint conditions that the structure does not fail and the sensing covers the entire area to be monitored. Solve the composite laminate optimization model using the adaptive genetic algorithm to obtain the optimal ply design scheme of the composite laminate. Optimize the embedded optical fiber composite material according to the optimal ply design scheme of the composite laminate and the optimal layout of the optical fiber sensor.
2. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 1, wherein The sensor layout optimization model is: where S j is a binary variable indicating whether the j-th candidate point is embedded with an optical fiber sensor. When the j-th candidate point is embedded with an optical fiber sensor, S j takes the value of 1. When the j-th candidate point is not embedded with an optical fiber sensor, S j takes the value of 0; J is the number of candidate points; Y is a vector composed of the layout parameters of the optical fiber sensors, and y i is the layout parameter of the i-th optical fiber sensor, n is the number of optical fiber sensors, and POD is the monitoring coverage rate of the optical fiber sensors.
3. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 1, wherein The determination process of the constraint condition for sensing to cover the entire area to be monitored includes: Select multiple points to be monitored on the composite material; the points to be monitored are ideal position points for embedding optical fiber sensors. Using the formula calculate the monitoring probability of each point to be monitored by a fiber optic sensor; in the formula, (x i , y i ) is the coordinate of the i-th fiber optic sensor, (x k , y k ) is the coordinate of the k-th point to be monitored, P(x k , y k , x i , y i ) is the monitoring probability of the k-th point to be monitored by the i-th fiber optic sensor, R is the distance between the node where the i-th fiber optic sensor is located and the k-th point to be monitored, R1 is the first distance threshold, and R2 is the second distance threshold; According to the monitoring probability of each point to be monitored by a fiber optic sensor, use the formula to determine the monitoring probability of each point to be monitored by a sensing system composed of multiple fiber optic sensors; in the formula, P(x k ,y k ) is the monitoring probability of the k-th point to be monitored by a sensing system composed of multiple fiber optic sensors, and m' is the number of fiber optic sensors in the sensing system; According to the monitoring probability of each point to be monitored by a sensing system composed of multiple fiber optic sensors, the average value of the monitoring probabilities of all points to be monitored by the sensing system composed of multiple fiber optic sensors is used as the global monitoring probability. With the goal of achieving a 100% coverage rate of all points to be monitored, the constraint condition for the sensing to cover the entire area to be monitored is In the formula, POD is the monitoring coverage rate of the fiber optic sensor, and n is the number of fiber optic sensors.
4. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 1, wherein The adaptive crossover operator is: Wherein, P c is the crossover probability, f' is the larger fitness value among the two individuals participating in crossover, f avg is the average fitness of the parent generation, f max is the maximum fitness of the parent generation, iter is the current iteration number, iter max is the maximum iteration number; The adaptive mutation operator is: where P m is the mutation probability, f is the fitness of the offspring individuals participating in the mutation operation, and P c1 , P c2 , P c3 , P m1 , P m2 , P m3 are all adjustable parameters.
5. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 4, characterized in that P c1 = 0.9, P c2 = 0.6, P c3 = 0.2, P m1 = 0.05, P m2 = 0.01, P m3 = 0.01。 6. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 1, characterized in that The adaptive genetic algorithm adopts an elitist strategy to select the best individuals as the initial population. The adaptive genetic algorithm adopts an evaluation function E in multiplicative form v = F v × P v , and transforms the constrained problem into an unconstrained problem; in the formula, E v is the evaluation function, F v is the fitness, and P v is the penalty function.
7. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 1, characterized in that The composite laminate optimization model is: where \(X\) is a vector composed of the ply parameters of the composite laminate, \(x\) l is the parameter of the \(l\)-th ply in the composite laminate, \(\omega_1\) and \(\omega_2\) are both weight functions, is the dimensionless weight of the embedded optical fiber composite material, is the dimensionless displacement of the embedded optical fiber composite material, \(I\) F is the failure factor, POD is the monitoring coverage rate of the fiber optic sensor, \(n\) is the number of fiber optic sensors, \(n\) min is the minimum number in the optimal layout of the fiber optic sensors, \(m\) is the weight of the embedded optical fiber composite material, \(u\) is the displacement of the embedded optical fiber composite material, \(m_0\) is the weight of the embedded optical fiber composite material in the initial design, and \(u_0\) is the displacement of the embedded optical fiber composite material in the initial design.
8. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 7, wherein, The determination process of the failure factor includes: Considering the influence of the optical fiber sensor on the mechanical properties of the structure, construct a finite element model of the composite laminate. Solve the finite element model of the composite laminate to obtain the mechanical response of the composite material. Based on the mechanical response of the composite material and the Tsai-Wu failure criterion, the failure factor is calculated using the formula ; where F 11 , F 22 , F 66 , F1, F2, F 12 are all strength parameters, σ 11 is the stress of the composite material in the 1 direction, σ 22 is the stress of the composite material in the 2 direction, and τ 12 is the shear stress of the composite material; Among them, X t is the axial or longitudinal tensile ultimate stress, X c is the axial or longitudinal compressive ultimate stress, Y t is the transverse tensile ultimate stress, Y c is the transverse compressive ultimate stress, and S is the shear ultimate stress.
9. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 1, characterized in that The ply parameters of the composite laminate include: ply angle, ply sequence, and ply thickness.
10. The optimization method of the embedded optical fiber composite material based on the adaptive genetic algorithm according to claim 9, characterized in that Use integer coding to describe the ply angle of the composite laminate, and the integer coding corresponds to the plies from the inside to the outside from left to right.
Citation Information
Cited By
Optical fiber embedding parameter optimization method for composite material laying layer defect detection
CN120524714A
A method for optimizing optical fiber embedding parameters for composite material ply defect detection
CN120524714B
