Gradient Lattice Structure Generation and Optimization Method and Device for Laser Additive Manufacturing

Abstract

The present invention belongs to the technical field of additive manufacturing, and discloses a method and device for generating and optimizing a gradient lattice structure for laser additive manufacturing, including: collecting design requirement data; generating a gradient lattice structure by using an improved evolutionary algorithm according to the design requirement data; evaluating whether the generated gradient lattice structure meets the standards; if the gradient lattice structure does not meet the standards, optimizing the gradient lattice structure; the present invention solves the problem that the existing method depends on manual modeling, and realizes the multi-objective optimization of the gradient lattice structure, effectively generates excellent gradient lattice structures that meet complex design requirements, can shorten the design cycle, improve product performance, and realize more efficient gradient lattice structure design in the field of laser additive manufacturing.

Description

Technical Field

[0001] The present invention relates to the field of additive manufacturing technology, and more specifically, to a method and device for generating and optimizing a gradient lattice structure for laser additive manufacturing. Background Art

[0002] With the continuous progress of manufacturing technology, laser additive manufacturing (LAM), as an emerging manufacturing method, has gradually received wide attention; this technology has the characteristics of high manufacturing flexibility, high material utilization rate, and large design freedom, and can effectively meet the needs of modern industry for complex gradient lattice structures and personalized products; in many applications, the gradient lattice structure has become a research hotspot due to its excellent mechanical properties and lightweight characteristics; the gradient lattice structure can effectively optimize the performance of products and improve the usage efficiency by realizing different physical and mechanical properties in the material; however, how to efficiently generate and optimize these complex gradient lattice structures during the laser additive manufacturing process still faces many challenges; existing methods often lack the flexibility and accuracy for specific applications and are difficult to achieve ideal gradient lattice structure performance and manufacturing efficiency.

[0003] For example, Chinese Patent with publication number CN111451505A discloses a selective laser melting preparation process for a variable density gradient material of a metal lattice structure; it includes: selecting a metal spherical powder suitable for selective laser melting as the raw material; using three-dimensional modeling software to construct the required variable density gradient lattice structure model, slicing the model with special software and then importing it into the selective laser melting forming equipment; setting appropriate selective laser melting forming process parameters, and the entire forming process is carried out in an argon or nitrogen atmosphere; after processing, the lattice structure together with the substrate is subjected to annealing treatment, and after annealing, it is wire-cut, surface-cleaned, and sandblasted to obtain; the variable density gradient lattice structure obtained by this invention has good surface quality and high specific strength, and while ensuring the functionality and reliability of the gradient lattice structure, it also meets the lightweight requirements of the gradient lattice structure; although the above method can realize the generation of the gradient lattice structure, through the inventor's research and practical application of the above method and the existing technology, it is found that the above method and the existing technology have at least the following partial defects:

[0004] (1) It depends on manual modeling and is limited to a certain extent in terms of complexity and flexibility; for gradient lattice structures with complex gradient changes or non-linear gradients, manual modeling is very difficult and time-consuming.

[0005] (2) It is only applicable to limited optimization goals (such as strength, lightweight), and it is very difficult to consider multiple complex performance indicators (such as mechanical properties, materials, shapes, etc.) at the same time, and multi-objective optimization is difficult.

[0006] (3) If the performance of the artificially designed gradient lattice structure fails to meet the standards, it is necessary to manually adjust the design again, increasing the design cycle.

[0007] In view of this, the present invention proposes a method and device for generating and optimizing gradient lattice structures for laser additive manufacturing to solve the above problems. Summary of the Invention

[0008] To overcome the above-mentioned defects of the prior art and achieve the above object, the present invention provides the following technical solutions: A method for generating and optimizing gradient lattice structures for laser additive manufacturing, including:

[0009] Collect design requirement data;

[0010] Generate a gradient lattice structure using an improved evolutionary algorithm according to the design requirement data;

[0011] Evaluate whether the generated gradient lattice structure meets the standards;

[0012] If the gradient lattice structure does not meet the standards, discretize the gradient lattice structure into multiple finite element units, initially set the density of each unit to the maximum value; with the goal of minimizing the structural compliance, and set the volume of the unit not to exceed the maximum allowable volume; iteratively adjust the unit density value by analyzing the sensitivity, optimize the distribution by combining smoothing processing, and repeatedly check whether the volume constraint is satisfied; if not, readjust the density value until the result converges to determine the optimized gradient lattice structure.

[0013] Further, the steps of generating a gradient lattice structure using an improved clonal selection algorithm include:

[0014] Step 1: Collect m structural samples, set different numerical labels for different structural samples, and mark them as structural labels, where m is an integer greater than 1;

[0015] Step 2: Randomly generate an initial antibody population, and the antibodies in the initial antibody population correspond one-to-one with the structural labels;

[0016] Step 3: Determine the fitness function;

[0017] Step 4: Calculate the fitness of each antibody in the initial antibody population, and perform antibody screening to generate a memory antibody population;

[0018] Step 5: Calculate the corresponding cloning scale for each antibody in the memory antibody population, and perform cloning to generate a cloned antibody population;

[0019] Step 6: Calculate the selection probability of each antibody in the cloned antibody population, and determine whether to select the antibody according to the selection probability to generate a selected antibody population;

[0020] Step 7: Perform cloud adaptive mutation on each antibody in the selected antibody group using the cloud adaptive mutation operator to generate a mutated antibody group;

[0021] Step 8: Generate new antibodies for the antibodies in the mutated antibody group using the interpolation method to generate an interpolation antibody group;

[0022] Step 9: Recombine the antibodies in the mutated antibody group to generate a recombined antibody group;

[0023] Step 10: Combine the interpolation antibody group and the recombined antibody group to generate a combined antibody group;

[0024] Step 11: Determine whether the iteration ends; if so, calculate the fitness of each antibody in the combined antibody group and obtain the structure sample corresponding to the structure label with the maximum fitness; if not, perform antibody screening on the combined antibody group, regenerate the memory antibody group, and return to Step 5.

[0025] Furthermore, in Step 1, the structure sample is the gradient lattice structure sample;

[0026] In Step 2, the initial antibody group , is the q-th antibody, that is, the scale of the initial antibody group is q; the range of antibodies is m structure samples; the iteration times t of the initial antibody population is 0;

[0027] In Step 4, the method of antibody screening is: screen out antibodies with the maximum fitness from the initial antibody group, ;

[0028] In Step 5, the calculation of the cloning scale is obtained by combining the fitness of the antibody and the affinity of the antibody, and the calculation is as follows:

[0029] ;

[0030] ;

[0031] In the formula, is the cloning scale of the i-th antibody, Int is the ceiling function, q is the scale of the initial antibody group, is the affinity of the i-th antibody, is the fitness of the i-th antibody, min is the minimum value function, exp is the natural exponential function, is the Euclidean distance between the i-th antibody and the j-th antibody, , .

[0032] Furthermore, the design requirement data includes additive manufacturing objects, mechanical requirement data, material requirement data, shape requirement data, and lightweight requirement data; the additive manufacturing objects include product type, functional requirements, and usage environment; the mechanical requirement data includes stress distribution, load conditions, and stiffness and flexibility requirements; the material requirement data includes material type and material distribution; the shape requirement data includes geometric dimensions, geometric shapes, and local density gradients; the lightweight requirement data includes density distribution and weight distribution;

[0033] In step 3, the expression of the fitness function is: ; where is the fitness, is the requirement satisfaction degree; the method for obtaining the requirement satisfaction degree is as follows: for the data in the design requirement data that is not a numerical value, different digital tags are set respectively and marked as requirement tags, and the digital tags of different data are all different; replace the data in the design requirement data that is not a numerical value with the corresponding requirement tags, and mark the replaced design requirement data as replaced data; mark the replaced data and the structure tags corresponding to the antibodies as analysis data, and input the analysis data into the trained requirement prediction model to predict the corresponding requirement satisfaction degree.

[0034] Furthermore, in step 6, the method for generating the selected antibody population includes:

[0035] Calculate the fitness of each antibody in the cloned antibody population and add them up in turn to obtain the total antibody degree. Divide the fitness of each antibody by the total antibody degree to obtain the selection probability of each antibody; sort the selection probabilities of each antibody from largest to smallest, and replace each selection probability with the sum of each selection probability and all the selection probabilities before the corresponding selection probability in positive order; obtain all the replaced selection probabilities and mark them as replaced probabilities; the antibody corresponding to the replaced probability is the same as the antibody corresponding to the selection probability before replacement; take every two adjacent replaced probabilities as an analysis set in positive order; mark the larger replaced probability in each analysis set as the first probability and the smaller replaced probability as the second probability; construct the probability screening range corresponding to each analysis set according to the first probability and the second probability of each analysis set, where the first probability is the maximum value of the probability screening range, the second probability is the minimum value of the probability screening range, and the minimum value is an open interval and the maximum value is a closed interval; each analysis set corresponds to the antibody corresponding to the second probability; among them, the minimum value of the probability screening range corresponding to the antibody ranked first is 0, and the maximum value is the corresponding selection probability; randomly generate a random number between , and select the antibodies corresponding to the probability screening ranges where the random numbers are located to generate the selected antibody population;

[0036] In step 8, the method for generating the interpolation antibody population includes:

[0037] Calculate the fitness of each antibody in the mutant antibody population, sort them from largest to smallest to generate a sorting table; screen out half of the antibodies in the mutant antibody population in ascending order according to the sorting table and label them as screened antibodies; generate new antibodies by using the interpolation method for every two adjacent screened antibodies; calculate the fitness corresponding to each new antibody and add it to the sorting table, and retain the antibodies corresponding to the top y fitness values in the sorting table to generate the interpolation antibody population, where y is the number of antibodies in the mutant antibody population;

[0038] The new antibody is obtained by weighted fusion of two adjacent screened antibodies with random weights, and the expression of the new antibody is: ; in the formula, is the new antibody, , are two adjacent screened antibodies, is the interpolation coefficient, is a random number between.

[0039] Further, in step 9, the method for antibody recombination includes:

[0040] Randomly select Y antibodies from the mutant antibody population as parental antibodies for antibody recombination to generate offspring antibodies , until the number of unselected antibodies in the mutant antibody population is less than Y, the antibody recombination is completed;

[0041] The offspring antibody is generated by the weighted average of Y randomly selected parental antibodies, and the calculation method includes:

[0042] ;

[0043] In the formula, is the Yth parental antibody, is the Yth proportionality factor, and the proportionality factor is a randomly generated real number and not all of the Y proportionality factors are 0;

[0044] In step 11, the method for determining whether the iteration ends includes:

[0045] A preset iteration threshold; compare the iteration number t with the iteration threshold; if the iteration number t is greater than or equal to the iteration threshold, the iteration ends; if the iteration number t is less than the iteration threshold, mark the largest fitness in the combined antibody population as the single maximum fitness; take the absolute value of the difference between the single maximum fitness in the current iteration process and the single maximum fitness in the previous iteration process as the fitness change amount;

[0046] Preset variable threshold and quantity threshold; compare the fitness change amount with the variable threshold. If the fitness change amount is greater than or equal to the variable threshold, no smoothing instruction is generated. If the fitness change amount is less than the variable threshold, a smoothing instruction is generated. Count the number of generated smoothing instructions and compare it with the quantity threshold. If the number of smoothing instructions is less than the quantity threshold, the iteration continues, and let the iteration count , and select the one with the maximum fitness in the combined antibody population antibodies to form a new memory antibody population. If the number of smoothing instructions is greater than or equal to the quantity threshold, the iteration ends.

[0047] Furthermore, the method for evaluating whether the generated gradient lattice structure meets the standard includes:

[0048] Perform finite element analysis on the generated gradient lattice structure using finite element software to obtain corresponding performance data. The performance data includes mechanical data and lightweight data, where the data in the mechanical data is consistent with the data in the mechanical requirement data, and the data in the lightweight data is consistent with the data in the lightweight requirement data. Use the performance data, mechanical requirement data, and lightweight requirement data as test data, input the test data into the trained structure evaluation model, output an evaluation label, obtain the corresponding evaluation result according to the evaluation label, and evaluate whether the generated gradient lattice structure meets the standard according to the evaluation result. The evaluation label is the digital label corresponding to the evaluation result, and the evaluation result includes meeting the standard and not meeting the standard, and the digital labels corresponding to different evaluation results are different. Train the structure evaluation model based on the test data and the evaluation label;

[0049] The training process of the structure evaluation model includes:

[0050] Convert the performance data, mechanical requirement data, and lightweight data into numerical features to ensure that the MLP can process continuous inputs, collect training samples, with the quantity ≥ 1000. Convert the judgment result "meeting the standard / not meeting the standard" into a binary classification label (0 / 1). Divide the training samples into a training set and a test set according to 7:3.

[0051] Use the training set to train the structure evaluation model. The structure evaluation model takes the evaluation label corresponding to each group of test data as the output, takes the actual evaluation label corresponding to each group of test data as the prediction target, and the actual evaluation label is the evaluation label pre-collected corresponding to the test data. Evaluate the model accuracy of the prediction result using the mean absolute percentage error MAPE. When the calculated MAPE is less than the preset MAPE, the training of the structure evaluation model is completed, and a structure evaluation model for predicting the evaluation label according to the test data is generated. Among them, the structure evaluation model is a deep belief network model.

[0052] Furthermore, the steps for optimizing the gradient lattice structure include:

[0053] Step a: Discretize the gradient lattice structure into finite element units, obtain the volume of each unit, and set an initial density value of 1 for each unit. ; ;

[0054] Step b: Define the objective function and volume constraint.

[0055] Step c: Calculate the sensitivity of the current gradient lattice structure.

[0056] Step d: Update the density value of each unit , and perform a smoothing operation.

[0057] Step e: After calculating the updated density value, calculate the volume of each unit in the current gradient lattice structure, and determine whether the volume constraint is satisfied. If not, adjust the density value of each unit, and then enter Step f. If satisfied, directly enter Step f. ;

[0058] Step f: Calculate the sensitivity of the current gradient lattice structure, and determine whether the optimization is completed. If not, return to Step d. If so, obtain the current gradient lattice structure as the optimized gradient lattice structure.

[0059] In the above Step b, the expression of the objective function is: ; where is the compliance, is the stiffness matrix, is the displacement vector, is the transpose of the displacement vector;

[0060] The expression of the volume constraint is: ; where is the volume of the gradient lattice structure, is the volume of the th unit, is the maximum allowable volume of the gradient lattice structure, , and M is the number of units in the gradient lattice structure.

[0061] Furthermore, in the above Step c, the method for calculating the sensitivity of the current gradient lattice structure includes: using the finite element method to calculate the displacement vector and stiffness matrix of the current gradient lattice structure, substituting them into the objective function, calculating the corresponding compliance, and calculating the sensitivity based on the compliance.

[0062] The expression of the sensitivity is: ; where is the sensitivity, is the penalty factor, is the compliance of the th unit;

[0063] In step d, update the density value of each unit The method is: Where, For the During the optimization process The density value of the unit, is the gradient coefficient, For the The sensitivity of each unit, ;

[0064] In step e, the method for determining whether the volume constraint is satisfied is as follows: substituting the volume of each unit in the current gradient lattice structure after the density value is updated into the volume constraint expression; if the volume constraint expression is established, the volume constraint is satisfied; if the volume constraint expression is not established, the volume constraint is not satisfied;

[0065] The density value for each cell The adjustment is done by calculating the scaling factor , scaling factor The expression is: ; The density value of each unit Multiply by the scaling factor , get the adjusted density value of each unit ;

[0066] In step f, the method for determining whether the optimization is completed is: presetting a change threshold; subtracting the sensitivity change calculated in the previous optimization process from the sensitivity change calculated in the current optimization process to obtain the change difference; if the change difference is less than the change threshold, the optimization is completed; if the change difference is greater than or equal to the change threshold, the optimization continues.

[0067] Furthermore, the step of obtaining the gradient coefficient includes:

[0068] Step d1: Preset the initial coefficient search interval ; Set the split ratio ;

[0069] Step d2: According to the segmentation ratio , the initial coefficient search interval Divide two new points and , , ;

[0070] Step d3: According to and Update the density value of each unit. After obtaining the displacement vector and stiffness matrix of the gradient lattice structure respectively after the unit density value is updated, calculate the corresponding compliance respectively; According to The compliance calculated after updating the unit density value is marked as the first compliance, and according to The compliance calculated after updating the unit density value is marked as the second compliance;

[0071] Step d4: If the first compliance is greater than or equal to the second compliance, the minimum value of the updated coefficient search interval is the point corresponding to the second compliance, and the maximum value is the maximum value of the coefficient search interval before update; If the first compliance is less than the second compliance, the minimum value of the updated coefficient search interval is the minimum value of the coefficient search interval before update, and the maximum value is the point corresponding to the first compliance;

[0072] Step d5: Preset a width threshold, subtract the minimum value from the maximum value in the updated coefficient search interval to obtain the interval width; Compare the interval width with the width threshold; If the interval width is greater than or equal to the width threshold, divide the updated coefficient search area according to the segmentation ratio to re-divide into two new points and , and return to step d3; If the interval width is less than the width threshold, take the average of the maximum value and the minimum value in the updated coefficient search area as the gradient coefficient.

[0073] The device for generating and optimizing the gradient lattice structure for laser additive manufacturing, implementing the method for generating and optimizing the gradient lattice structure for laser additive manufacturing, includes:

[0074] A data acquisition module for collecting design requirement data;

[0075] A structure generation module that generates a gradient lattice structure using an improved evolutionary algorithm according to the design requirement data;

[0076] A structure evaluation module for evaluating whether the generated gradient lattice structure meets the standards;

[0077] A structure optimization module for optimizing the gradient lattice structure if it does not meet the standards.

[0078] The technical effects and advantages of the method and device for generating and optimizing the gradient lattice structure for laser additive manufacturing of the present invention:

[0079] According to the collected design requirement data, an improved evolutionary algorithm is used to automatically generate a gradient lattice structure, realizing the automatic conversion of the design requirement data into a gradient lattice structure, and solving the problem that the existing methods rely on manual modeling. At the same time, multiple design indicators are considered to realize the multi-objective optimization of the gradient lattice structure, which can improve the performance of the gradient lattice structure in multiple aspects and effectively generate an excellent gradient lattice structure that meets complex design requirements. In addition, if the generated gradient lattice structure does not meet the standards, the gradient lattice structure can be automatically optimized, which can shorten the design cycle and improve the product performance, thus realizing a more efficient design of the gradient lattice structure in the field of laser additive manufacturing. BRIEF DESCRIPTION OF THE DRAWINGS

[0080] Figure 1 FIG. is a schematic diagram of a device for generating and optimizing a gradient lattice structure for laser additive manufacturing according to Embodiment 1 of the present invention;

[0081] Figure 2 FIG. is a flowchart of a method for generating a gradient lattice structure according to Embodiment 1 of the present invention;

[0082] Figure 3 FIG. is a flowchart of a method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to Embodiment 2 of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0083] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0084] Embodiment 1

[0085] Please refer to Figure 1 As shown in the figure, the device for generating and optimizing a gradient lattice structure for laser additive manufacturing in this embodiment includes a data acquisition module, a structure generation module, a structure evaluation module, and a structure optimization module; each module is connected by wired and / or wireless means to realize data transmission between modules;

[0086] The data acquisition module is used to acquire design requirement data.

[0087] The design requirement data at least includes an additive manufacturing object, mechanical requirement data, material requirement data, shape requirement data, and lightweight requirement data; the design requirement data is obtained by manual input by relevant staff.

[0088] The additive manufacturing object at least includes product type, functional requirements, usage environment, etc.; among which, product types such as medical devices (such as implants, prosthetics, dental restorations, etc.), aerospace components (such as engine components, brackets, casings, etc.), automotive parts (such as lightweight gradient lattice structures, components with complex shapes, etc.); functional requirements such as load-bearing components (such as frames, supports, etc.), thermal management components (such as radiators, heat exchangers, etc.); usage environments such as high-temperature environments, corrosive environments, impact environments, etc.

[0089] The mechanical requirement data at least includes stress distribution, load conditions, stiffness and flexibility requirements, etc.; among which, the stress distribution is the stress field information that the gradient lattice structure bears in the working environment, usually including stress concentration or stress gradient in different regions; the load conditions are the external loads (such as static loads, dynamic loads, impact loads, etc.) that the gradient lattice structure bears during use, including working conditions such as tension, compression, and bending; the stiffness and flexibility requirements are the requirements for stiffness or flexibility in different regions. For example, in lightweight design, some regions require high stiffness, while other regions may require better flexibility.

[0090] The material requirement data at least includes material type, material distribution, etc.; among which, the material type is the metal or alloy powder material used in the gradient lattice structure, such as aluminum alloy, stainless steel, titanium alloy, etc.; different material types have different melting points, thermal conductivities, and mechanical properties, which directly affect the performance of the gradient lattice structure; the material distribution is the regional distribution of different materials in the gradient lattice structure.

[0091] The shape requirement data at least includes geometric dimensions, geometric shapes, local density gradients, etc.; among which, the geometric dimensions are the overall dimensions of the gradient lattice structure; the geometric shape is the unit geometric shape of the gradient lattice structure, such as tetrahedron, hexahedron, honeycomb, etc. Different geometric shapes have different mechanical properties, which affect the stress transfer, stiffness, and energy absorption capacity of the gradient lattice structure; the local density gradient is the degree of change in density within a local area of the gradient lattice structure.

[0092] The lightweight requirement data at least includes density distribution and weight distribution, etc.; among which, the density distribution is the density change of the material within the gradient lattice structure; the weight distribution is the weight distribution of different regions within the gradient lattice structure; through lightweight design, the density and weight distribution can be optimized to reduce material consumption while ensuring the load-bearing capacity of the gradient lattice structure, thereby reducing the manufacturing cost.

[0093] It should be noted that the reason for collecting design requirement data is as follows: The additive manufacturing object defines the type of gradient lattice structure for additive manufacturing, such as aerospace components, medical implants, or mechanical parts, etc.; the manufacturing object defines the overall design goal and affects the functional requirements of the gradient lattice structure (such as strength, stiffness, elasticity, etc.); the mechanical requirement data covers information such as the mechanical stress, load, and stress distribution that the manufacturing object needs to withstand; based on the mechanical requirement data, the gradient lattice structure can be optimized to adapt to the load requirements of different regions; for example, regions with high stress require a denser lattice, while regions with low stress can use a more lightweight gradient lattice structure; the material requirement data affects the mechanical properties, thermal conductivity, corrosion resistance, etc. of the gradient lattice structure. Different materials have different performances under conditions such as stress and temperature. It is necessary to generate the optimal gradient lattice structure according to the material characteristics to ensure the stability and performance of the gradient lattice structure in a specific environment; the shape requirement data of the additive manufacturing object directly affects the design of the gradient lattice structure. If the additive manufacturing object has a complex geometric shape, the gradient lattice structure needs to be locally adjusted in different regions to ensure adaptation to the boundary. For example, the lattice units in the curved surface region may require specific arrangements and density changes; through the lightweight requirement data, the gradient lattice structure can be optimized to minimize the material usage while maintaining the strength, achieving the goal of lightweighting.

[0094] The structure generation module generates a gradient lattice structure according to the design requirement data by using an improved evolutionary algorithm.

[0095] Evolutionary algorithms such as genetic algorithms, clonal selection algorithms, etc. In this embodiment, taking clonal selection as an example, an improved clonal selection algorithm is used to generate a gradient lattice structure.

[0096] As Figure 2 shown, the steps of generating a gradient lattice structure by using an improved clonal selection algorithm include:

[0097] Step 1: Collect m structural samples, set different numerical labels for different structural samples, and mark them as structural labels, where m is an integer greater than 1;

[0098] Step 2: Randomly generate an initial antibody population, and the antibodies in the initial antibody population correspond one-to-one with the structural labels;

[0099] Step 3: Determine the fitness function;

[0100] Step 4: Calculate the fitness of each antibody in the initial antibody population, and perform antibody screening to generate a memory antibody population;

[0101] Step 5: Calculate the corresponding cloning scale for each antibody in the memory antibody population, and perform cloning to generate a cloned antibody population;

[0102] Step 6: Calculate the selection probability of each antibody in the cloned antibody group, and determine whether to select the antibody based on the selection probability to generate a selected antibody group;

[0103] Step 7: Perform cloud adaptive mutation on each antibody in the selected antibody group using a cloud adaptive mutation operator to generate a mutant antibody group. Cloud adaptive mutation is an existing technology and will not be described in detail here.

[0104] Step 8: Using an interpolation method to generate new antibodies for the antibodies in the variant antibody group to generate an interpolated antibody group;

[0105] Step 9: performing antibody recombination on the antibodies in the variant antibody group to generate a recombinant antibody group;

[0106] Step 10: Merge the interpolated antibody group and the recombinant antibody group to generate a merged antibody group;

[0107] Step 11: Determine whether the iteration is complete; if so, calculate the fitness of each antibody in the combined antibody group, and obtain the structural sample corresponding to the structural label of the antibody with the largest fitness; if not, perform antibody screening on the combined antibody group, regenerate the memory antibody group, and return to step 5.

[0108] In the above step 1, the structure sample is a gradient lattice structure sample, which is obtained by technicians in this field by collecting gradient lattice structures generated in historical laser additive manufacturing processes, or extracted from an existing laser additive manufacturing database.

[0109] In step 2 above, the initial antibody population , is the qth antibody, that is, the size of the initial antibody population is q; the range of antibodies is m structural samples; the number of iterations t of the initial antibody population is 0.

[0110] In step 3 above, the fitness function is expressed as: Where, For fitness, is the demand satisfaction; the method for obtaining the demand satisfaction is: for the data that are not numerical values in the design demand data (such as product type, usage environment, material type, etc.), different digital labels are set respectively, and marked as demand labels, wherein the digital labels of different data are different; for example, the digital label of the prosthesis in the product type is set to 1, the digital label of the bracket is set to 2, the digital label of the aluminum alloy in the material type is set to 100, and the digital label of the stainless steel is set to 101; the data that are not numerical values in the design demand data are replaced with the corresponding demand labels, and the replaced design demand data are marked as replacement data; the replacement data and the structural labels corresponding to the antibody are marked as analysis data, and the analysis data is input into the trained demand prediction model to predict the corresponding demand satisfaction.

[0111] The specific training process of the demand prediction model includes:

[0112] Pre-collect the demand satisfaction degrees corresponding to b groups of analysis data, where b is an integer greater than 1, and convert the analysis data and the corresponding demand satisfaction degrees into a corresponding set of feature vectors; the demand satisfaction degrees corresponding to the analysis data are collected by those skilled in the art during the generation process of the historical gradient lattice structure. b groups of different analysis data are collected, and experimental analysis is carried out sequentially under the conditions of each group of analysis data. The demand satisfaction degree of the generated gradient lattice structure for the corresponding design requirement data is analyzed, that is, the demand satisfaction degree corresponding to each group of analysis data is obtained, and corresponding demand satisfaction degrees are set for the b groups of analysis data respectively.

[0113] Use the demand satisfaction degrees corresponding to the b groups of analysis data as the training data of the demand prediction model.

[0114] Data preprocessing, including normalizing (such as standardizing) the replacement data to ensure consistent feature distributions. One-hot encoding or label encoding is performed on the categorical data of the antibody structure labels. The demand satisfaction degree is numerical data and is directly used as the target value.

[0115] Model selection, select linear regression, random forest, neural network, etc. Taking the neural network as an example.

[0116] Input layer: Combine the replacement data and the antibody structure labels; Hidden layer: Use the ReLU activation function to capture non-linear relationships; Output layer: A single neuron outputs the demand satisfaction degree.

[0117] Model training, divide the training data into a training set (80%) and a validation set (20%); Use the mean squared error (MSE) loss function and the Adam optimizer to train the model. Monitor the performance of the validation set to prevent overfitting.

[0118] Model evaluation, use the test set to verify the model, calculate the mean squared error MSE, and stop training until the sum of the mean squared errors reaches convergence.

[0119] In step 4 above, the method for antibody screening is: Screen out the antibody with the maximum fitness from the initial antibody population, .

[0120] In step 5 above, the calculation of the cloning scale is obtained by combining the fitness of the antibody and the affinity of the antibody, and the calculation is as follows:

[0121] ;

[0122] ;

[0123] In the formula, $n_{i}$ is the cloning scale of the $i$-th antibody, Int is the ceiling function to ensure that the cloning scale is an integer, meeting the requirements of practical applications, and $q$ is the scale of the initial antibody population. $A_{i}$ is the affinity of the $i$-th antibody. By calculating the minimum distance between antibodies, the diversity of the population is measured. If an antibody is very different from other antibodies in the population ( large), its cloning scale is increased to avoid premature convergence of the population to a local optimum; if antibodies are too similar ( small), cloning is inhibited to prevent redundancy. $F_{i}$ is the fitness of the $i$-th antibody. Fitness dominates "exploitation" (focusing on the current high-quality solutions), and affinity dominates "exploration" (encouraging diversity). The combination of the two ensures that the algorithm achieves a balance between global search and local optimization; min is the minimum value function, and exp is the natural exponential function. $d_{ij}$ is the Euclidean distance between the $i$-th antibody and the $j$-th antibody. , .

[0124] It should be noted that when calculating the Euclidean distance between the $i$-th antibody and the $j$-th antibody, normalization processing is performed to make .

[0125] The calculation of the cloning scale is based on two core logics. First, high-quality individuals are screened through the fitness ratio of antibodies, ensuring that the higher the fitness of an antibody, the greater the probability of being replicated, thereby retaining dominant genes. Second, the concept of affinity is introduced to measure the difference between an antibody and other individuals in the population. Specifically, the Euclidean distance between an antibody and its nearest neighbor antibody is calculated and exponentially amplified. If the antibody is highly unique (large distance), its cloning quantity is increased; otherwise, redundancy is inhibited. This design combines the "exploitation" and "exploration" mechanisms: fitness drives the algorithm to converge to the current high-quality solutions, while affinity actively maintains the diversity of the population and avoids falling into a local optimum. At the same time, the cloning scale is converted to an integer by rounding up to ensure the feasibility of actual operations. The overall strategy dynamically balances the global search ability and local development efficiency during the optimization process.

[0126] In the above step 6, the method for generating the selected antibody population includes:

[0127] Calculate the fitness of each antibody in the cloned antibody population, sum them up in sequence to obtain the total antibody degree, divide the fitness of each antibody by the total antibody degree respectively to obtain the selection probability of each antibody; sort the selection probabilities of each antibody from large to small, and replace each selection probability with the sum of each selection probability and all the selection probabilities before the corresponding selection probability in positive order; obtain all the replaced selection probabilities and mark them as replacement probabilities; the antibody corresponding to the replacement probability is the same as the antibody corresponding to the selection probability before replacement; take every two adjacent replacement probabilities as an analysis set in positive order; mark the larger replacement probability in each analysis set as the first probability and the smaller replacement probability as the second probability; construct the probability screening range corresponding to each analysis set according to the first probability and the second probability of each analysis set, where the first probability is the maximum value of the probability screening range, the second probability is the minimum value of the probability screening range, and the minimum value is an open interval and the maximum value is a closed interval; each analysis set corresponds to the antibody corresponding to the second probability; among them, the minimum value of the probability screening range corresponding to the antibody ranked first is 0 and the maximum value is the corresponding selection probability; randomly generate a random number between and select the antibody corresponding to the probability screening range where the random number is located to generate a selected antibody population.

[0128] Exemplarily, the selection probabilities of three antibodies are 0.5, 0.3, and 0.2 in sequence; since there is no value before 0.5, 0.5 remains 0.5 after replacement; since there is 0.5 before 0.3, 0.3 is replaced with 0.3 + 0.5 = 0.8, and since there are 0.3 and 0.5 before 0.2, 0.2 is replaced with 0.2 + 0.3 + 0.5 = 1; therefore, the replacement probabilities are 0.5, 0.8, and 1 in sequence; 0.5 and 0.8 form an analysis set, so the probability screening range corresponding to the second antibody is ; 0.8 and 1 also form an analysis set, so the probability screening range corresponding to the third antibody is ; the probability screening range corresponding to the first antibody is ; if the randomly generated random number is 0.7, since 0.7 is within , the second antibody is selected.

[0129] In the above step 8, the method for generating an interpolated antibody population includes:

[0130] Calculate the fitness of each antibody in the mutant antibody population, sort them from largest to smallest, and generate a sorting table; screen out half of the antibodies in the mutant antibody population in ascending order according to the sorting table and label them as screened antibodies; generate new antibodies for every two adjacent screened antibodies using the interpolation method; calculate the fitness corresponding to each new antibody and add it to the sorting table, and retain the antibodies corresponding to the top y fitness values in the sorting table to generate an interpolation antibody population, where y is the number of antibodies in the mutant antibody population.

[0131] The new antibody is obtained by weighted fusion of two adjacent screened antibodies with random weights, and the expression of the new antibody is: ; In the formula, is the new antibody, 、 are two adjacent screened antibodies, is the interpolation coefficient, is a random number between.

[0132] When = 0, the new antibody completely inherits the characteristics of ; when = 1, the new antibody completely inherits the characteristics of ; the intermediate value mixes the characteristics of both in proportion.

[0133] Generate new antibodies from the selected high-quality antibodies ( 、 ), ensuring that the algorithm performs a fine search near the high-quality solution and improving the local optimization ability. The random number introduces uncertainty, making the generation position of the new antibody randomly distributed between the two, avoiding the population from converging to a single region prematurely; by mixing the characteristics of different antibodies, the new antibody not only retains the dominant genes of the parent generation but also generates certain mutations, increasing the population diversity and preventing from falling into local optima; the linear interpolation operation is simple and has low computational cost, suitable for large-scale iterative optimization scenarios, while ensuring the generation efficiency; on the basis of retaining the local development ability of the high-quality solution, expand the search range through random interpolation, taking into account both the convergence speed and the global search potential of the algorithm.

[0134] In step 9 above, the method of antibody recombination includes:

[0135] Randomly select Y antibodies from the mutant antibody population as parent antibodies for antibody recombination to generate offspring antibodies , until the number of unselected antibodies in the mutant antibody population is less than Y, the antibody recombination is completed; in this embodiment, Y is preferably 3; the recombinant antibody population includes multiple offspring antibodies .

[0136] Offspring antibody It is generated by the weighted average of Y randomly selected parent antibodies, specifically as follows:

[0137] ;

[0138] In the formula, is the Y-th parent antibody, is the Y-th scaling factor. The scaling factor is a randomly generated real number and not all of the Y scaling factors are 0, ensuring the randomness of the generation process of the offspring antibodies, avoiding the population from falling into local optima, and enhancing the global search ability; by fusing the characteristics of multiple parent antibodies through weighted averaging, the offspring inherit the advantages of different parents, balancing exploration (new characteristics) and exploitation (parent advantages); the denominator is the sum of the scaling factors, preventing the weights from being too large or too small, resulting in numerical instability, and ensuring that the calculation result is a reasonable weighted average.

[0139] In step 11 above, the methods for determining whether the iteration ends include:

[0140] Preset an iteration threshold, which is preset by those skilled in the art according to the algorithm accuracy;

[0141] Compare the iteration number t with the iteration threshold;

[0142] If the iteration number t is greater than or equal to the iteration threshold, the iteration ends;

[0143] If the iteration number t is less than the iteration threshold, mark the maximum fitness in the combined antibody population as the single maximum fitness; take the absolute value of the difference between the single maximum fitness in the current iteration process and the single maximum fitness in the previous iteration process as the fitness change amount;

[0144] Preset a variable threshold and a quantity threshold, both of which are preset by those skilled in the art according to the actual situation; compare the fitness change amount with the variable threshold. If the fitness change amount is greater than or equal to the variable threshold, no smoothing instruction is generated; if the fitness change amount is less than the variable threshold, a smoothing instruction is generated; count the number of generated smoothing instructions and compare it with the quantity threshold;

[0145] If the number of smoothing instructions is less than the quantity threshold, the iteration continues, and let the iteration number , and select the antibodies with the maximum fitness in the combined antibody population to form a new memory antibody population; where is a computer language, that is, add 1 to the value of the iteration number t and then assign it to the iteration number t;

[0146] If the number of smoothing instructions is greater than or equal to the quantity threshold, the iteration ends.

[0147] It should be noted that the reason for using the improved clonal selection algorithm to generate the gradient lattice structure is as follows:

[0148] 1. Performance optimization: The improved clonal selection algorithm improves the efficiency of the optimization process by simulating the principle of the biological immune system; it can effectively find the optimal solution and is suitable for complex gradient lattice structure optimization problems;

[0149] 2. Handling multi-objective optimization: In laser additive manufacturing, design requirements often involve multiple performance indicators (such as mechanical properties, materials, lightweight, etc.); the improved clonal selection algorithm can consider multiple objectives simultaneously and adapt to complex design requirements;

[0150] 3. Enhanced flexibility: The improved clonal selection algorithm can explore more flexibly in the search space by introducing individual cloning and mutation mechanisms, adapting to different gradient changes and non-linear gradient structure designs;

[0151] 4. Reducing the need for manual modeling: The improved clonal selection algorithm can automatically generate and optimize structures, reducing the complexity and time cost brought by human factors;

[0152] 5. Strong adaptability: The clonal selection algorithm can dynamically adjust parameters according to the design requirement data, making the generated gradient lattice structure more in line with the actual usage requirements and improving the performance and reliability of the product.

[0153] 6. Accelerating the iterative process: Through adaptive mutation and antibody recombination, the improved clonal selection algorithm can accelerate the convergence speed, improve the optimization efficiency, and shorten the design cycle.

[0154] A structure evaluation module for evaluating whether the generated gradient lattice structure meets the standards.

[0155] The methods for evaluating whether the generated gradient lattice structure meets the standards include:

[0156] Performing finite element analysis on the generated gradient lattice structure using finite element software (such as ANSYS, Abaqus, etc.) to obtain corresponding performance data; the performance data includes at least mechanical data and lightweight data, where the data in the mechanical data is consistent with the data in the mechanical requirement data, and the data in the lightweight data is consistent with the data in the lightweight requirement data;

[0157] Take the performance data, mechanical requirement data, and lightweight requirement data as test data, input the test data into the trained structure evaluation model, output the evaluation label, obtain the corresponding evaluation result according to the evaluation label, and evaluate whether the generated gradient lattice structure meets the standard according to the evaluation result; the evaluation label is the digital label corresponding to the evaluation result, and the evaluation result includes meeting the standard and not meeting the standard, and the digital labels corresponding to different evaluation results are different. Exemplarily, set the digital label for meeting the standard to 0.1 and the digital label for not meeting the standard to 0.2.

[0158] The specific training process of the structure evaluation model includes:

[0159] Pre-collect group A training samples. Each group of training samples includes test data and the corresponding judgment result. A is an integer greater than 1. The judgment result corresponding to the test data is collected by those skilled in the art for A groups of different test data during the historical generation process of the gradient lattice structure. Those skilled in the art compare the performance data, corresponding mechanical requirement data, and lightweight requirement data in the A groups of different test data according to actual experience, judge whether the corresponding gradient lattice structure meets the standard, and set the corresponding judgment results for the A groups of different test data in sequence.

[0160] Divide the training samples into a training set and a test set according to 7:3.

[0161] Use the training set to train the structure evaluation model. The structure evaluation model takes the evaluation label corresponding to each group of test data as the output and the actual evaluation label corresponding to each group of test data as the prediction target. The actual evaluation label is the evaluation label pre-collected corresponding to the test data; use the mean absolute percentage error MAPE to evaluate the model accuracy for the prediction result. When the calculated MAPE is less than the preset MAPE, the structure evaluation model training is completed, and a structure evaluation model for predicting the evaluation label according to the test data is generated; among them, the calculation formula of MAPE is ; among them, is the actual evaluation label corresponding to the d-th group of test data, is the predicted evaluation label corresponding to the d-th group of test data, d is the group number of the feature vector corresponding to the test data, and A represents the number of predicted evaluation labels, ; generate a structure evaluation model for predicting the evaluation label according to the test data; among them, the structure evaluation model is a deep belief network model, and the preset MAPE is pre-set by those skilled in the art according to the accuracy required by the structure evaluation model.

[0162] The structure optimization module is used to optimize the gradient lattice structure if it does not meet the standard.

[0163] The steps for optimizing the gradient lattice structure include:

[0164] Step a: Discretize the gradient lattice structure into finite element units using finite element software, obtain the volume of each unit, and set an initial density value of 1 for each unit, which represents the distribution state of the material. The finite element software will automatically calculate and store the volume of each unit, so it can be directly read. For 1; , representing the distribution state of the material; the finite element software will automatically calculate and store the volume of each unit, so it can be directly read.

[0165] Step b: Define the objective function and volume constraint.

[0166] The expression of the objective function is: ; In the formula, is the compliance, is the stiffness matrix, and the stiffness matrix changes with the density value. is the displacement vector, is the transpose of the displacement vector; among them, both the displacement vector and the stiffness matrix are obtained by analyzing the gradient lattice structure using the finite element method. The finite element method is an existing technology and will not be elaborated here. The compliance is the deformation ability of the gradient lattice structure under the action of external forces. Specifically, the compliance can be defined as the relationship between the load borne by the gradient lattice structure and the corresponding displacement, and is usually used to evaluate the stiffness characteristics of the gradient lattice structure. The smaller the compliance, the greater the stiffness of the gradient lattice structure.

[0167] The expression of the volume constraint is: ; In the formula, is the volume of the gradient lattice structure, is the volume of the th unit, is the maximum allowable volume of the gradient lattice structure, which is set according to the actual shape requirements (such as lightweight target or load-bearing capacity) to ensure that the design meets both functional requirements and resource limitations. , M is the number of units in the gradient lattice structure; among them, the volume of the gradient lattice structure is also obtained by analyzing the gradient lattice structure using the finite element method, and the maximum allowable volume of the gradient lattice structure is obtained through the shape requirement data.

[0168] It should be noted that in the volume constraint expression, dividing by is for the purpose of normalizing the volume constraint, that is, converting the volume constraint into a relative standard quantity to facilitate the control and management of the material distribution in the optimization process of the gradient lattice structure.

[0169] The volume constraint formula restricts the material usage of the gradient lattice structure by normalizing the weighted volume sum: the left side of the formula represents the sum of the products of the volumes of all units and their densities (or weights) , and then divided by the total volume , ensure that this ratio does not exceed the maximum allowable volume . Its core logic is to control the distribution or total amount of materials within the structure, avoid excessive consumption of resources, and at the same time, through normalization, make the constraint conditions independent of the structure size, only focusing on the reasonable allocation of volume.

[0170] Step c: Calculate the sensitivity of the current gradient lattice structure. The sensitivity reflects the impact of the change in density value on the overall structural compliance in each unit.

[0171] The methods for calculating the sensitivity of the current gradient lattice structure include:

[0172] Use the finite element method to calculate the displacement vector and stiffness matrix of the current gradient lattice structure, substitute them into the objective function, calculate the corresponding compliance, and calculate the sensitivity based on the compliance;

[0173] The expression for sensitivity is: ; In the formula, is the sensitivity, is the penalty factor, usually taking a value of 2 or 3, is the th unit compliance. The negative sign in the formula indicates that regions with high compliance (i.e., units with softer structures or larger deformations) will significantly reduce the total sensitivity, thus guiding the design variables to adjust in the direction of reducing compliance (enhancing stiffness) during the optimization process, meeting the requirements of structural lightweight and toughening.

[0174] When = 2, will amplify the sensitivity of low-density or medium-density regions, driving the design variables towards 0 or 1, thus eliminating the fuzzy boundary and accelerating convergence to a clear structure. Control of non-linear effects: By adjusting the value (usually taking 2 or 3), balance the sensitivity sensitivity to different density regions, and avoid the optimization process falling into local extrema.

[0175] Compliance reflects the deformation ability of the unit under force. High compliance corresponds to low stiffness. By incorporating it into the sensitivity calculation, the formula directly correlates the structural performance (stiffness) with the design variables, ensuring the achievement of the optimization goal (such as minimizing compliance).

[0176] Step d: Update the density value of each unit , and perform a smoothing operation to ensure continuous change of density values within the neighborhood and avoid numerical instability problems; The smoothing operation is an existing technology and will not be elaborated here.

[0177] Update the density value of each unit The methods include:

[0178] ;

[0179] In the formula, is the density value of the th unit in the th optimization process, is the gradient coefficient, is the th unit sensitivity, .

[0180] The steps for obtaining the gradient coefficient include:

[0181] Step d1: Preset the initial coefficient search interval , preferably in this embodiment, ; Set the segmentation ratio , preferably in this embodiment;

[0182] Step d2: According to the segmentation ratio , divide the initial coefficient search interval into two new points and , , ;

[0183] Step d3: Respectively update the density value of each unit according to and . After the unit density value is updated respectively, obtain the displacement vector and stiffness matrix of the gradient lattice structure, and calculate the corresponding compliance respectively; Mark the compliance calculated after the unit density value is updated according to as the first compliance, and mark the compliance calculated after the unit density value is updated according to as the second compliance;

[0184] Step d4: If the first compliance is greater than or equal to the second compliance, the minimum value of the updated coefficient search interval is the point corresponding to the second compliance, and the maximum value is the maximum value of the coefficient search interval before the update; If the first compliance is less than the second compliance, the minimum value of the updated coefficient search interval is the minimum value of the coefficient search interval before the update, and the maximum value is the point corresponding to the first compliance; Exemplarily, after the initial coefficient search interval passes through the points and and is updated, if the first compliance is greater than or equal to the second compliance, the updated coefficient search interval is , if the first compliance is less than the second compliance, the updated coefficient search interval is ;

[0185] Step d5: Preset a width threshold, which is preset by those skilled in the art according to the actual situation. Subtract the minimum value from the maximum value in the updated coefficient search interval to obtain the interval width. Compare the interval width with the width threshold. If the interval width is greater than or equal to the width threshold, re-divide the updated coefficient search area according to the segmentation ratio , and re-divide into two new points and , and return to step d3; if the interval width is less than the width threshold, take the average value of the maximum value and the minimum value in the updated coefficient search area as the gradient coefficient.

[0186] Step e: After calculating the updated density value, calculate the volume of each unit in the current gradient lattice structure, and determine whether the volume constraint is satisfied. If not, adjust the density value of each unit , and then enter step f. If it is satisfied, directly enter step f; it should be understood that when the density value of each unit in the gradient lattice structure changes, the gradient lattice structure will also change accordingly. Therefore, the volume of each unit in the current gradient lattice structure will also change accordingly.

[0187] The method for determining whether the volume constraint is satisfied includes:

[0188] Substitute the volume of each unit in the current gradient lattice structure after updating the density value into the expression of the volume constraint. If the expression of the volume constraint holds, the volume constraint is satisfied, that is ; if the expression of the volume constraint does not hold, the volume constraint is not satisfied, that is .

[0189] The method for adjusting the density value of each unit includes:

[0190] Calculate the scaling factor , and the expression of the scaling factor is: ; multiply the density value of each unit by the scaling factor respectively to obtain the adjusted density value of each unit .

[0191] Step f: Calculate the sensitivity of the current gradient lattice structure, and determine whether the optimization is completed. If not, return to step d; if so, obtain the current gradient lattice structure as the optimized gradient lattice structure.

[0192] The method for determining whether the optimization is completed includes:

[0193] A preset change amount threshold, which is preset by those skilled in the art according to the actual situation; subtract the sensitivity change amount calculated in the previous optimization process from the sensitivity change amount calculated in the current optimization process to obtain a change amount difference; compare the change amount difference with the change amount threshold; if the change amount difference is less than the change amount threshold, the optimization ends, and if the change amount difference is greater than or equal to the change amount threshold, the optimization continues.

[0194] It should be noted that the purpose of optimizing the gradient lattice structure by using the above steps a - f is as follows:

[0195] 1. Improve the structural performance: Through the optimization process, it aims to reduce the compliance and increase the structural stiffness, so that the gradient lattice structure has better load-bearing capacity and stability under external forces;

[0196] 2. Meet the design requirements: Ensure that the optimized structure can meet specific mechanical requirements, material requirements and lightweight goals, and ensure the reliability and efficiency of the gradient lattice structure in practical applications.

[0197] 3. Achieve efficient utilization of materials: Through controlling the density distribution and volume constraints, the optimization process can achieve the effective utilization of materials, reduce unnecessary material waste, and thus reduce the manufacturing cost;

[0198] 4. Enhance the adaptability of the structure: Through sensitivity analysis, the optimization process can adjust the material distribution according to different loads and environmental conditions, making the structure have stronger adaptability.

[0199] 5. Improve production efficiency: Through optimized design, it can reduce the complexity in subsequent processing and manufacturing, thereby improving the overall production efficiency and reducing costs.

[0200] In this embodiment, according to the collected design requirement data, an improved evolutionary algorithm is used to automatically generate a gradient lattice structure, realizing the automatic conversion of the design requirement data into a gradient lattice structure, and solving the problem that the existing method relies on manual modeling; at the same time, considering multiple design indicators, multi-objective optimization of the gradient lattice structure is realized, which can improve the performance of the gradient lattice structure in many aspects and effectively generate an excellent gradient lattice structure that meets complex design requirements; in addition, if the generated gradient lattice structure does not meet the standards, the gradient lattice structure can be automatically optimized, which can shorten the design cycle and improve the product performance, so as to achieve a more efficient gradient lattice structure design in the field of laser additive manufacturing.

[0201] Embodiment 2

[0202] Please refer to Figure 3 As shown, for the parts not described in detail in this embodiment, please refer to the description content of Embodiment 1. A method for generating and optimizing a gradient lattice structure for laser additive manufacturing is provided, and the method includes:

[0203] Collect design requirement data;

[0204] Generate a gradient lattice structure by using an improved evolutionary algorithm according to the design requirement data;

[0205] Evaluate whether the generated gradient lattice structure meets the standards;

[0206] If the gradient lattice structure does not meet the standards, discretize the gradient lattice structure into multiple finite element units, and initially set the density of each unit to the maximum value; aim to minimize the structural compliance, and set that the volume of the unit does not exceed the maximum allowable volume; iteratively adjust the unit density value by analyzing the sensitivity, and optimize the distribution by combining smoothing processing, and repeatedly check whether the volume constraint is satisfied; if not, readjust the density value until the result converges to determine the optimized gradient lattice structure.

[0207] As described above, it is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present invention can easily think of changes or substitutions, which should be covered by the protection scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

[0208] Finally: The above are only the preferred embodiments of the present invention and are not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for generating and optimizing a gradient lattice structure for laser additive manufacturing, characterized in that, Including: Collecting design requirement data; The design requirement data includes an additive manufacturing object, mechanical requirement data, material requirement data, shape requirement data, and lightweight requirement data; the additive manufacturing object includes product type, functional requirements, and usage environment; the mechanical requirement data includes stress distribution, load conditions, and stiffness and flexibility requirements; the material requirement data includes material type and material distribution; the shape requirement data includes geometric dimensions, geometric shape, and local density gradient; the lightweight requirement data includes density distribution and weight distribution; Generating a gradient lattice structure using an improved evolutionary algorithm according to the design requirement data; Evaluating whether the generated gradient lattice structure meets the standards; If the gradient lattice structure does not meet the standards, discretize the gradient lattice structure into multiple finite element units, and initially set the density of each unit to the maximum value; aiming to minimize the structural compliance and setting the volume of the unit not to exceed the maximum allowable volume; iteratively adjust the unit density value by analyzing the sensitivity, optimize the distribution by combining smoothing processing, and repeatedly check whether the volume constraint is satisfied; if not, readjust the density value until the result converges to determine the optimized gradient lattice structure.

2. The method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to claim 1, characterized in that The steps of generating a gradient lattice structure using an improved clonal selection algorithm include: Step 1: Collect m structural samples, set different numerical labels for different structural samples, and mark them as structural labels, where m is an integer greater than 1; Step 2: Randomly generate an initial antibody population, and the antibodies in the initial antibody population correspond one-to-one with the structural labels; Step 3: Determine the fitness function; Step 4: Calculate the fitness of each antibody in the initial antibody population, and perform antibody screening to generate a memory antibody population; Step 5: Calculate the corresponding cloning scale for each antibody in the memory antibody population, and perform cloning to generate a cloned antibody population; Step 6: Calculate the selection probability of each antibody in the cloned antibody population, and determine whether to select the antibody according to the selection probability to generate a selected antibody population; Step 7: Perform cloud adaptive mutation on each antibody in the selected antibody population using a cloud adaptive mutation operator to generate a mutant antibody population; Step 8: Generate new antibodies for the antibodies in the mutant antibody population using interpolation method to generate an interpolation antibody population; Step 9: Recombine the antibodies in the mutant antibody population to generate a recombinant antibody population; Step 10: Combine the interpolation antibody population and the recombinant antibody population to generate a combined antibody population; Step 11: Determine whether the iteration ends; if so, calculate the fitness of each antibody in the combined antibody population, and obtain the structural sample corresponding to the structural label of the antibody with the maximum fitness; if not, perform antibody screening on the combined antibody population, regenerate the memory antibody population, and return to Step 5.

3. The method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to claim 2, wherein In Step 1, the structural sample is a gradient lattice structure sample; In the said step 2, the initial antibody population , is the q-th antibody, that is, the scale of the initial antibody population is q; the range of the antibody is m structural samples; the iteration number t of the initial antibody population is 0; In the said step 4, the method for antibody screening is as follows: screen out antibodies with the maximum fitness from the initial antibody population, ; In Step 5, the cloning scale is calculated by combining the fitness of the antibody and the affinity of the antibody; 4. The method for generating and optimizing the gradient lattice structure for laser additive manufacturing according to claim 3, wherein In step 3, the expression of the fitness function is as follows: ; where is the fitness, is the requirement satisfaction degree; the method for obtaining the requirement satisfaction degree is as follows: for the data in the design requirement data that are not numerical values, different numerical labels are respectively set and marked as requirement labels, and the numerical labels of different data are all different; the data in the design requirement data that are not numerical values are replaced with the corresponding requirement labels, and the replaced design requirement data are marked as replaced data; the replaced data and the structure labels corresponding to the antibodies are marked as analysis data, and the analysis data are input into the trained requirement prediction model to predict the corresponding requirement satisfaction degree.

5. The method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to claim 4, wherein In Step 6, the method for generating a selected antibody population includes: Calculate the fitness of each antibody in the cloned antibody population, add them up in sequence to obtain the total antibody degree, divide the fitness of each antibody by the total antibody degree respectively to obtain the selection probability of each antibody; sort the selection probabilities of each antibody from large to small, and sequentially replace each selection probability with the sum of each selection probability and all the selection probabilities before the corresponding selection probability according to the positive order; obtain all the replaced selection probabilities and mark them as replacement probabilities; the antibody corresponding to the replacement probability is the same as the antibody corresponding to the selection probability before replacement; take every two adjacent replacement probabilities as an analysis set according to the positive order; mark the larger replacement probability in each analysis set as the first probability and the smaller replacement probability as the second probability; construct the probability screening range corresponding to each analysis set according to the first probability and the second probability of each analysis set, where the first probability is the maximum value of the probability screening range, the second probability is the minimum value of the probability screening range, and the minimum value is an open interval and the maximum value is a closed interval; each analysis set corresponds to the antibody corresponding to the second probability; among them, the minimum value of the probability screening range corresponding to the antibody ranked first is 0 and the maximum value is the corresponding selection probability; randomly generate pieces random numbers between, select the antibody corresponding to the probability screening range where the random numbers are located to generate the selected antibody population; In Step 8, the method for generating an interpolation antibody population includes: Calculate the fitness of each antibody in the mutant antibody population, sort them from largest to smallest to generate a sorting table; screen out half of the antibodies in the mutant antibody population in ascending order according to the sorting table and label them as screened antibodies; generate new antibodies for every two adjacent screened antibodies using the interpolation method; calculate the fitness corresponding to each new antibody and add it to the sorting table, and retain the antibodies corresponding to the top y fitness values in the sorting table to generate an interpolation antibody population, where y is the number of antibodies in the mutant antibody population; The new antibody is obtained by weighted fusion of two adjacent screened antibodies with random weights.

6. The method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to claim 5, characterized in that In step 9, the method for antibody recombination includes: Randomly select Y antibodies from the mutant antibody population as parental antibodies for antibody recombination to generate offspring antibodies Antibody recombination is completed until the number of unselected antibodies in the mutant antibody population is less than Y; Offspring antibody Generated by the weighted average of Y randomly selected parent antibodies; In step 11, the method for determining whether the iteration ends includes: Preset an iteration threshold; compare the iteration number t with the iteration threshold; if the iteration number t is greater than or equal to the iteration threshold, the iteration ends; if the iteration number t is less than the iteration threshold, mark the largest fitness value in the combined antibody population as the single maximum fitness value; take the absolute value of the difference between the single maximum fitness value in the current iteration process and the single maximum fitness value in the previous iteration process as the fitness change amount; Preset variable threshold and quantity threshold; compare the fitness change amount with the variable threshold. If the fitness change amount is greater than or equal to the variable threshold, no smoothing instruction is generated; if the fitness change amount is less than the variable threshold, a smoothing instruction is generated; count the number of generated smoothing instructions and compare it with the quantity threshold; if the number of smoothing instructions is less than the quantity threshold, the iteration continues, and the iteration count , and select the one with the maximum fitness in the combined antibody population antibodies to form a new memory antibody population; if the number of smoothing instructions is greater than or equal to the quantity threshold, the iteration ends.

7. The method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to claim 1, wherein The method for evaluating whether the generated gradient lattice structure meets the standard includes: Perform finite element analysis on the generated gradient lattice structure using finite element software to obtain corresponding performance data; the performance data includes mechanical data and lightweight data, where the data in the mechanical data is consistent with the data in the mechanical requirement data, and the data in the lightweight data is consistent with the data in the lightweight requirement data; use the performance data, mechanical requirement data, and lightweight requirement data as test data, input the test data into the trained structure evaluation model, output an evaluation label, obtain the corresponding evaluation result according to the evaluation label, and evaluate whether the generated gradient lattice structure meets the standard according to the evaluation result; the evaluation label is the digital label corresponding to the evaluation result, and the evaluation results include meeting the standard and not meeting the standard, and the digital labels corresponding to different evaluation results are different.

8. The method for generating and optimizing the gradient lattice structure for laser additive manufacturing according to claim 7, wherein The steps for optimizing the gradient lattice structure include: Step a: Discretize the gradient lattice structure into finite - element units, obtain the volume of each unit, and set an initial density value of 1 for each unit. ; Step b: Define the objective function and volume constraint; Step c: Calculate the sensitivity of the current gradient lattice structure; Step d: Update the density value of each unit ; and perform a smoothing operation; Step e: After updating the density value, calculate the volume of each unit in the current gradient lattice structure, and determine whether the volume constraint is satisfied. If not, adjust the density value of each unit , and then enter Step f. If satisfied, directly enter Step f; ​ Step f: Calculate the sensitivity of the current gradient lattice structure, determine whether the optimization is completed, if not, return to step d; if so, obtain the current gradient lattice structure as the optimized gradient lattice structure; In the said step b, the expression of the objective function is: ; in the formula, is compliance, is the stiffness matrix, is the displacement vector, is the transpose of the displacement vector; The expression for the volume constraint is: ; where is the volume of the gradient lattice structure, is the volume of the th unit, is the maximum allowable volume of the gradient lattice structure, , and M is the number of units in the gradient lattice structure.

9. The method for generating and optimizing the gradient lattice structure for laser additive manufacturing according to claim 8, wherein In step c, the method for calculating the sensitivity of the current gradient lattice structure includes: using the finite element method to calculate the displacement vector and stiffness matrix of the current gradient lattice structure, and substituting them into the objective function to calculate the corresponding compliance, and calculating the sensitivity according to the compliance; The expression for the sensitivity is as follows: ; where is the sensitivity, is the penalty factor, is the compliance of the th element; In step d, update the density value of each unit The method is as follows: ; where is the density value of the th unit in the th optimization process, is the gradient coefficient, is the sensitivity of the th unit, ; In step e, the method for determining whether the volume constraint is satisfied is: substitute the volume of each unit in the current gradient lattice structure after updating the density value into the expression of the volume constraint. If the expression of the volume constraint holds, the volume constraint is satisfied; if the expression of the volume constraint does not hold, the volume constraint is not satisfied; The density value of each unit The method for adjustment is: calculate the scaling factor , and the scaling factor has the expression: ; multiply the density value of each unit respectively by the scaling factor to obtain the adjusted density value of each unit ; In step f, the method for determining whether the optimization is completed is: preset a change amount threshold; subtract the sensitivity change amount calculated in the previous optimization process from the sensitivity change amount calculated in the current optimization process to obtain a change amount difference; if the change amount difference is less than the change amount threshold, the optimization is completed, and if the change amount difference is greater than or equal to the change amount threshold, the optimization continues.

10. The method for generating and optimizing the gradient lattice structure for laser additive manufacturing according to claim 9, wherein The steps for obtaining the gradient coefficient include: Step d1: Preset the initial coefficient search interval ; Set the splitting ratio ; Step d2: According to the segmentation ratio , divide the initial coefficient search interval into two new points and , , ; Step d3: respectively update the density value of each unit according to and After the density value of the unit is updated, respectively obtain the displacement vector and stiffness matrix of the gradient lattice structure, and calculate the corresponding compliance respectively; Mark the compliance calculated after the unit density value is updated according to as the first compliance, and mark the compliance calculated after the unit density value is updated according to as the second compliance; Step d4: If the first compliance is greater than or equal to the second compliance, the minimum value of the updated coefficient search interval is the point corresponding to the second compliance, and the maximum value is the maximum value of the coefficient search interval before the update; if the first compliance is less than the second compliance, the minimum value of the updated coefficient search interval is the minimum value of the coefficient search interval before the update, and the maximum value is the point corresponding to the first compliance; Step d5: Preset a width threshold, subtract the minimum value from the maximum value in the updated coefficient search interval to obtain the interval width; compare the interval width with the width threshold; if the interval width is greater than or equal to the width threshold, re-divide the updated coefficient search area according to the segmentation ratio , and re-divide it into two new points and , and return to step d3; if the interval width is less than the width threshold, use the mean value of the maximum and minimum values in the updated coefficient search area as the gradient coefficient.

11. A device for generating and optimizing a gradient lattice structure for laser additive manufacturing, which implements the method for generating and optimizing a gradient lattice structure for laser additive manufacturing according to any one of claims 1-10, characterized in that, including: A data acquisition module for collecting design requirement data; A structure generation module that uses an improved evolutionary algorithm to generate a gradient lattice structure according to the design requirement data; A structure evaluation module for evaluating whether the generated gradient lattice structure meets the standards; A structure optimization module for optimizing the gradient lattice structure if it does not meet the standards.

Images