Aluminum alloy thermal deformation strength parameter optimization method and system
By using an adaptive neuro-fuzzy inference system and a multi-objective optimization framework, the problem of insufficient robustness in the hot deformation process of aluminum alloys was solved, achieving high-performance and high-stability process parameter optimization and improving the industrial application effect of aluminum alloys.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-11-26
- Publication Date
- 2026-04-07
AI Technical Summary
Existing technologies lack robustness in optimizing hot deformation processes for aluminum alloys, making it difficult to achieve a balance between high performance and high stability in industrial environments. Furthermore, they fail to effectively address uncertainties in process parameters and microstructure dynamics.
An adaptive neural fuzzy inference system and a multi-objective optimization framework are adopted. The uncertainty of thermal deformation is handled by the fuzzy inference engine, and the optimization direction is determined by the Pareto optimal solution set, so as to achieve the synergistic optimization of the thermal deformation strength and process robustness of aluminum alloy.
It significantly improves the robustness and stability of the hot deformation process parameters of aluminum alloys, ensuring a combination of high performance and high stability process parameters, and fully tapping the inherent performance potential of aluminum alloys.
Smart Images
Figure CN121809716A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of aluminum alloy processing technology, and more specifically, to a method and system for optimizing the hot deformation strength parameters of aluminum alloys. Background Technology
[0002] Aluminum alloys, due to their excellent properties such as lightweight, high strength, and corrosion resistance, play a crucial role in high-end equipment fields such as aerospace, automotive manufacturing, and rail transportation. Hot deformation processes, such as hot extrusion, hot rolling, and hot forging, are the core manufacturing steps that endow aluminum alloys with their final shape and properties. During hot deformation, the setting of process parameters such as temperature, strain rate, and deformation amount directly determines the evolution path of the microstructure within the material, including dislocation density, grain size, and precipitate morphology, thus profoundly affecting its final mechanical properties, especially strength. Therefore, precisely controlling the hot deformation process parameters to achieve optimized control of microstructure and macroscopic properties is a core objective pursued in the field of high-performance aluminum alloy manufacturing, and is of vital importance for improving product quality, performance consistency, and market competitiveness. Currently, existing technologies have some shortcomings in optimizing aluminum alloy hot deformation processes. On the one hand, traditional optimization methods often rely on deterministic models under ideal conditions. These models, when predicting material behavior, often ignore objective uncertainties such as equipment condition fluctuations, slight differences in material composition, and measurement errors in the industrial production environment, leading to deviations in theoretical optimization results in practical applications and insufficient process robustness. On the other hand, when performing multi-objective optimization, existing technologies not only struggle to systematically balance the two mutually restrictive objectives of "high strength" and "high stability," but also typically fail to deeply couple and integrate the complex micro-organism dynamics of dislocation evolution, dynamic recrystallization, and precipitation phases with the uncertainties of macroscopic process parameters, thus limiting the ultimate exploration of material performance potential and the precise control of process windows. Summary of the Invention
[0003] The purpose of this invention is to provide a method and system for optimizing the hot deformation strength parameters of aluminum alloys, which achieves synergistic optimization of the hot deformation strength and process robustness of aluminum alloys, and obtains process parameters that take into account both high performance and high stability.
[0004] This invention is achieved through the following technical solution: A method for optimizing the hot deformation strength parameters of aluminum alloys, the method comprising the following steps: Step S1: Obtain the initial parameters of the hot deformation of the aluminum alloy; Step S2: Solve for the thermal deformation strength compatibility of the aluminum alloy based on the initial parameters of the aluminum alloy's thermal deformation; Step S3: Based on the thermal deformation intensity coordination degree, input the fuzzy inference engine to solve the membership function. The fuzzy inference engine adopts an adaptive neural fuzzy inference system, which maps the thermal deformation intensity coordination degree to the membership value in the interval [0,1] and incorporates a multi-objective optimization framework to handle the uncertainty of thermal deformation. Step S4: Based on the output value of the membership function, the optimization direction is determined by combining the fuzzy rule base and the Pareto optimal solution set. If the output value is lower than the dynamic threshold, the initial parameters of thermal deformation are adaptively adjusted to improve the thermal deformation intensity coordination. If the output value reaches or exceeds the dynamic threshold, the initial parameters of thermal deformation are not adjusted. Step S5: Iteratively execute steps S2-S4 until the optimization objective is reached, obtain robust hot deformation strength parameters, and complete the optimization of aluminum alloy hot deformation strength parameters. The optimization objective is defined as maximizing the hot deformation strength coordination while minimizing the multi-objective optimization framework.
[0005] Optionally, the specific calculation steps for solving the compatibility of the hot deformation strength of the aluminum alloy are as follows: Temperature, strain rate, deformation, and initial dislocation density and initial grain size values from the initial parameters of aluminum alloy hot deformation are extracted as calculation inputs. The strain rate value is multiplied by the deformation value to obtain the strain increment. The current dislocation density value is obtained by balancing the dislocation hardening and recovery terms. The hardening term is a function of the strain increment, and the recovery term is a negative exponential function of the temperature value. The temperature and deformation values are substituted into an exponential function to calculate the conversion fraction of the precipitated phase from the initial state to the stable state. The conversion fraction accelerates with increasing temperature and accumulates with increasing deformation. The above results are aggregated by multidimensional fuzzy mapping. The dislocation density value, precipitation phase transformation fraction and initial grain size value are multiplied by dynamic weight coefficients based on deformation stage, and the weighted average is solved to obtain the thermal deformation strength compatibility.
[0006] Optionally, the fuzzy inference engine is constructed using the following steps: Define input and output variables, take the thermal deformation strength compatibility as the input variable, and the membership value in the [0,1] interval as the output variable. Integrate the objective function in the multi-objective optimization framework as an auxiliary input to handle uncertainty; The input variables are fuzzified based on Gaussian membership functions, where the input variables correspond to the first fuzzy set, the second fuzzy set, and the third fuzzy set, respectively, and the parameters of the Gaussian membership functions are initialized. A fuzzy rule base is constructed based on the physical mechanism of hot deformation of aluminum alloy and historical hot deformation test data. The fuzzy rule base is configured in Takagi-Sugeno form, and the fuzzy rules cover all combinations of input variables. Historical aluminum alloy hot deformation test data were acquired and used as a training set. Gaussian membership function parameters and fuzzy rule weights were adaptively adjusted using backpropagation algorithm and least squares method. Verify whether the training error is less than the preset threshold. If not, iteratively adjust the Gaussian membership function parameters. If yes, complete the training of the fuzzy inference engine.
[0007] Optionally, the multi-objective optimization framework specifically aims to minimize the composite value of the velocity exponential deviation, pressure logarithmic deviation, and thick-wall pressure exponential decay. The objective function is calculated as follows:
[0008] in, Let be the objective function. , , These are the adaptive weight coefficients, For flow velocity measurement point index, This represents the total number of flow velocity measurement points. For the first Flow velocity values at each measuring point The average flow velocity at all measuring points. For the velocity variance, For pressure measurement point index, This represents the total number of pressure measurement points. For the first Pressure values at each measuring point This is the average pressure at all measuring points. For thick-walled hydrostatic pressure, For the coordination component index, The number of coordination dimensions. For the first One coordination component.
[0009] Optionally, the constraints of the multi-objective optimization framework include parameter range constraints, microstructure constraints, performance threshold constraints, and uncertainty coordination constraints. The parameter range constraint is calculated using the following formula:
[0010] in, , These are the lower and upper limits of the temperature, respectively. This is the heat distortion temperature. , These are the lower and upper limits of the strain rate, respectively. For strain rate, , These are the lower and upper limits of the deformation, respectively. This is the amount of deformation; The microstructure constraint is calculated using the following formula:
[0011] in, For dislocation density, This is the upper limit of dislocation density. To restore the score, Grain size, This is the lower limit of the grain size. Used as a reference strain rate; The performance threshold constraint is calculated using the following formula:
[0012] in, For the standard deviation of flow rate, This represents the upper limit of the standard deviation of the flow rate. For pressure standard deviation, This represents the upper limit of the pressure standard deviation. This is the lower limit of hydrostatic pressure for thick-walled structures. The calculation formula for the uncertainty coordination constraint is as follows:
[0013] in, For the degree of uncertainty coordination, As the lower limit of the degree of coordination, For pressure variance, This represents the maximum permissible variance.
[0014] Optionally, the step of determining the optimization direction based on the output value of the membership function, combined with the fuzzy rule base and the Pareto optimal solution set, specifically involves: Obtain the output value of the membership function; The output value is used as input, along with the initial parameters of the current hot deformation of the aluminum alloy and the microstructure vector, and then input into the fuzzy rule base. The preliminary optimization direction vector is generated through fuzzy rule base reasoning, and the constraints of the objective function of the multi-objective optimization framework are incorporated. The Pareto optimal solution set is generated based on the NSGA-II algorithm. The initial optimization direction vector is mapped onto the Pareto solution set, the weighted distance of each solution is calculated, and the nearest Pareto solution is selected. The final optimization direction is determined. If the selected Pareto solution satisfies the convergence condition, that is, the output value of the membership function increases to the set value, the objective function of the multi-objective optimization framework decreases to the set value, and all constraints are satisfied, then the adjusted parameter set is output. The parameter set includes the adjusted temperature, strain rate, and deformation. Otherwise, the initial parameters of the thermal deformation are adaptively adjusted to improve the thermal deformation strength coordination.
[0015] Optionally, the optimization objective is defined as maximizing the thermal deformation strength compatibility while minimizing the multi-objective optimization framework, and its specific calculation formula is as follows:
[0016] in, To optimize the objective, For the thermal deformation strength compatibility, The penalty coefficient is... As an indicator of uncertainty, For indexing measurement points with uncertainties, The total number of measurement points with uncertainties. For the first Uncertainty rheological value at the measuring point For reference rheological values, For the first Stress value at measuring point For reference stress, For the index, The normalization coefficient is... Index for the deformation stage, This represents the number of deformation stages. For the first Phase robust weights For the first The rate constant of the precipitation phase in stages, For the first Phase Zener-Hollomon parameters, For the first Stage index parameters, For the first Stage deformation amount For the first Stage strain rate, For the first Stage dislocation density, For the first Stage grain size, For maximum dislocation density, This represents the upper limit of grain size. It is a micro-power exponent.
[0017] A system for optimizing the hot deformation strength parameters of aluminum alloys, comprising: Input module, to obtain initial parameters of thermal deformation of aluminum alloy; The calculation module solves for the thermal deformation strength compatibility of aluminum alloys based on the initial parameters of the aluminum alloy's hot deformation. The optimization module solves for the membership function based on the thermal deformation intensity coordination degree input to the fuzzy inference engine. The fuzzy inference engine adopts an adaptive neural fuzzy inference system, which maps the thermal deformation intensity coordination degree to membership values in the [0,1] interval and incorporates a multi-objective optimization framework to handle the uncertainty of thermal deformation. Based on the output value of the membership function, the optimization direction is determined by combining the fuzzy rule base and the Pareto optimal solution set. If the output value is lower than the dynamic threshold, the initial parameters of thermal deformation are adaptively adjusted to improve the thermal deformation intensity coordination degree. If the output value reaches or exceeds the dynamic threshold, the initial parameters of thermal deformation are not adjusted. The optimization calculation is iteratively executed until the optimization objective is reached, wherein the optimization objective is defined as maximizing the thermal deformation intensity coordination degree while minimizing the multi-objective optimization framework. The output module outputs robust thermal deformation strength parameters.
[0018] Optionally, the calculation module further includes a coordination degree solution submodule, used to extract temperature, strain rate, deformation, and initial dislocation density and initial grain size values from the initial parameters of aluminum alloy hot deformation as calculation inputs; multiply the strain rate by the deformation to obtain the strain increment; calculate the balance based on the dislocation hardening and recovery terms to obtain the current dislocation density value, where the hardening term is a function of the strain increment and the recovery term is a negative exponential function of the temperature; substitute the temperature and deformation values into the exponential function to calculate the transformation fraction of the precipitated phase from the initial state to the stable state, the transformation fraction accelerates with increasing temperature and accumulates with increasing deformation; aggregate the above results through multidimensional fuzzy mapping, multiply the dislocation density, precipitated phase transformation fraction, and initial grain size value by dynamic weighting coefficients based on the deformation stage, and solve the weighted average to obtain the hot deformation strength coordination degree.
[0019] Optionally, the output module further includes a visualization submodule for performing visualization of the output thermal deformation strength parameters.
[0020] The technical solution of the present invention has at least the following advantages and beneficial effects: This invention, on the one hand, constructs uncertainty coordination and hot deformation strength coordination, and introduces a multi-objective optimization function integrating a microstructure evolution model. This enables quantitative evaluation and proactive control of process uncertainties during hot deformation, significantly improving the robustness of optimized process parameters and ensuring the stability of industrial production. On the other hand, this invention not only achieves synergistic optimization of the two core objectives of hot deformation strength and process robustness through non-dominated sorting genetic algorithm (NSGA-II) and Pareto optimal solution set evaluation, but also, with the help of fuzzy inference mechanism, can handle global optimization problems under complex constraints within a unified framework, thereby obtaining the optimal combination of process parameters that balances high performance and high stability, fully exploring the inherent performance potential of aluminum alloys. Attached Figure Description
[0021] Figure 1 A flowchart illustrating the method for optimizing the hot deformation strength parameters of aluminum alloys provided by this invention; Figure 2 A schematic diagram of the principle of the aluminum alloy hot deformation strength parameter optimization system provided by the present invention. Detailed Implementation
[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. The components of the embodiments of the present invention described and shown in the accompanying drawings can generally be arranged and designed in various different configurations.
[0023] Example 1 This embodiment details a method for optimizing the hot deformation strength parameters of aluminum alloys. The method iteratively optimizes the initial hot deformation parameters through steps S1 to S5 to obtain robust hot deformation strength parameters. The core of this embodiment lies in solving for the hot deformation strength compatibility, constructing a fuzzy inference engine, incorporating a multi-objective optimization framework to handle uncertainties, and determining the optimization direction through the Pareto optimal solution set. The final optimization objective is defined as maximizing the hot deformation strength compatibility while minimizing the multi-objective optimization framework. This method is applicable to high-strength aluminum alloys in the aerospace field and can be verified through laboratory hot compression experiments or finite element simulations.
[0024] Step S1: Obtain the initial parameters of the thermal deformation of the aluminum alloy. In this embodiment, the initial parameters of the aluminum alloy's hot deformation are first collected or input, including temperature, strain rate, deformation, and the initial dislocation density and initial grain size in the initial microstructure vector. In actual operation, a hot compression testing machine can be used to obtain the initial data under laboratory conditions: assuming an initial temperature T = 400°C, strain rate... Deformation amount Initial dislocation density Initial grain size =50μm. The data can be obtained through real-time sensor measurement or import from a historical database, with the aim of providing fundamental data for solving the thermal deformation strength compatibility problem. If the parameters are incomplete, they can be supplemented using interpolation methods (such as linear interpolation), but must meet preset ranges, such as the lower temperature limit. =300°C, upper limit =500°C to avoid invalid input.
[0025] Step S2: Solve for the thermal deformation strength compatibility of the aluminum alloy based on the initial parameters of the aluminum alloy thermal deformation.
[0026] Based on the parameters obtained in step S1, the thermal deformation strength compatibility is calculated. This calculation step specifically includes the following sub-steps to ensure the coupling between microstructure evolution and macroscopic parameters: First, the temperature, strain rate, and deformation values, as well as the initial dislocation density and initial grain size values from the initial microstructure vector, are extracted from the initial parameters of thermal deformation and used as calculation inputs. For example, the above example values are used: T = 400°C. , , , =50μm.
[0027] Secondly, the strain rate value is multiplied by the deformation value to obtain the strain increment. The current dislocation density value is obtained by balancing the dislocation hardening and recovery terms. The hardening term is a function of the strain increment, such as... The recovery term is a negative exponential function of the temperature value, such as... ,in, To activate energy, Let be the gas constant. The calculation formula can be simplified to: ,in, and Let be a material constant. In the example, assume... =150kJ / mol, =8.314 J / mol·K, then .
[0028] Substituting the temperature and deformation values into an exponential function, the fractional transformation of the precipitated phase from the initial state to the stable state is calculated. The conversion fraction increases rapidly with increasing temperature and accumulates with increasing deformation. As an example, the conversion fraction can be expressed as... Where k is the rate constant, To release activation energy, =100kJ / mol). In the example... ≈0.7.
[0029] The above results are aggregated by multidimensional fuzzy mapping. The dislocation density value, precipitation phase transformation fraction, and initial grain size value are multiplied by dynamic weighting coefficients based on the deformation stage, and the weighted average is calculated to obtain the thermal deformation strength compatibility. The formula is: ,in, , This represents the current grain size, estimated using the Hall-Petch relation. In the example, ≈0.65. To ensure that the degree of coordination reflects the balance between strength and microstructure, fuzzy mapping can be implemented using MATLAB functions.
[0030] Step S3: Solve the membership function based on the thermal deformation strength compatibility degree input to the fuzzy inference engine.
[0031] The thermal deformation intensity coordination degree obtained in step S2 is used as input to the fuzzy inference engine to solve for the membership function. The fuzzy inference engine adopts an adaptive neural fuzzy inference system (ANFIS), which maps the thermal deformation intensity coordination degree to membership values in the interval [0,1] and incorporates a multi-objective optimization framework to handle the uncertainty of thermal deformation. The construction steps are as follows: Define input and output variables: As input variables, the membership values in the [0,1] interval are used as output variables. The objective function in the multi-objective optimization framework is used as an auxiliary input to handle uncertainty. For example, the auxiliary input includes the uncertainty coordination degree C.
[0032] The input variables are fuzzified using a Gaussian membership function, where the input variables correspond to a first fuzzy set (low consistency, e.g., 0-0.3), a second fuzzy set (medium consistency, e.g., 0.3-0.7), and a third fuzzy set (high consistency, e.g., 0.7-1), respectively. The Gaussian membership function parameters are initialized (mean μ=0.5, standard deviation σ=0.2). The Gaussian function form is as follows: .
[0033] A fuzzy rule base is constructed based on the physical mechanism of hot deformation of aluminum alloys and historical hot deformation test data. The fuzzy rule base is configured in Takagi-Sugeno form, and the fuzzy rules cover all combinations of input variables. For example, rule 1: If If the value is low, then the output membership value is 0.2 * +0.1. Assume there are 9 rules.
[0034] Historical aluminum alloy hot deformation test data were acquired and used as the training set. Gaussian membership function parameters and fuzzy rule weights were adaptively adjusted using backpropagation and least squares methods. The training process involved initializing weights, calculating the forward output, adjusting parameters via error backpropagation, and iterating 100 times.
[0035] To verify if the training error is less than a preset threshold, understandably, if mean squared error is used, the preset threshold is set to 0.01. If not, iteratively adjust the Gaussian membership function parameters; if yes, the training of the fuzzy inference engine is complete. In the example, the input... =0.65, output membership value≈0.75.
[0036] Step S4: Determine the optimization direction based on the output value of the membership function, combined with the fuzzy rule base and the Pareto optimal solution set.
[0037] Based on the membership function output value from step S3, the optimization direction is determined by combining the fuzzy rule base and the Pareto optimal solution set. Specific process: Obtain the output value of the membership function.
[0038] The output value, along with the initial parameters of the current hot deformation of the aluminum alloy and the microstructure vector, is input into the fuzzy rule base. The fuzzy rule base then infers and generates a preliminary optimization direction vector. For example, the vector might represent a temperature adjustment ΔT = +20°C and a strain rate adjustment... It also incorporates the constraints of the objective function within a multi-objective optimization framework.
[0039] The Pareto optimal solution set is generated based on the NSGA-II algorithm: the initial population size N=100, the number of generations G=50, the crossover rate 0.9, and the mutation rate 0.1. The objective is to minimize the multi-objective function Obj. The Pareto solution set includes non-dominated solutions, such as 10 solutions corresponding to different values T. , combination.
[0040] The initial optimization direction vector is mapped onto the Pareto solution set, the weighted distance of each solution is calculated, and the nearest Pareto solution is selected.
[0041] Determine the final optimization direction: If the selected Pareto solution satisfies the convergence condition, i.e., the output value of the membership function increases to a set value (e.g., Δ membership function output value > 0.1), the objective function of the multi-objective optimization framework decreases to a set value (e.g., ΔObj < -0.05), and all constraints are met, then output the adjusted parameter set. For example, after adjustment, T = 420°C. , Otherwise, the initial parameters of thermal deformation are adaptively adjusted to improve the coordination of thermal deformation strength, so as to ensure the robustness of the optimization direction and handle multi-objective conflicts through NSGA-II.
[0042] Step S5: Iterate through steps S2-S4 until the optimization objective is achieved, and obtain robust thermal deformation strength parameters.
[0043] Iteratively execute steps S2 to S4 until the optimization objective is achieved, thus completing the optimization of the aluminum alloy hot deformation strength parameters. The optimization objective is defined as maximizing the hot deformation strength compatibility while minimizing the multi-objective optimization framework. The specific calculation formula is as follows:
[0044] in, To optimize the objective, For the thermal deformation strength compatibility, The penalty coefficient is... =1, As an indicator of uncertainty, For indexing measurement points with uncertainties, The total number of measurement points with uncertainties. =10, For the first Uncertainty rheological value at the measuring point For reference rheological values, For the first Stress value at measuring point For reference stress, For the index, =2, The normalization coefficient is... =1, Index for the deformation stage, This represents the number of deformation stages. =3, For the first Phase robust weights =0.3, For the first The rate constant of the precipitation phase in stages, =0.1, For the first Phase Zener-Hollomon parameters, , For the first Stage index parameters, =0.5, For the first Stage deformation amount For the first Stage strain rate, For the first Stage dislocation density, For the first Stage grain size, For maximum dislocation density, = Upper limit of grain size, = 100 μm, For micro-power exponent, =0.5.
[0045] In the iteration, the initial =0.5, iterate 10 times, recalculate after each parameter update. and ,like If the value is greater than 0.8 (threshold), stop and output robustness parameters, such as T=450°C. , 0.85.
[0046] The multi-objective optimization framework specifically aims to minimize the composite value of the velocity exponential deviation, pressure logarithmic deviation, and thick-wall pressure exponential decay. The objective function is calculated as follows:
[0047] in, Let be the objective function. , , These are the adaptive weight coefficients, =0.4, =0.3, =0.3, For flow velocity measurement point index, This represents the total number of flow velocity measurement points. =20, For the first Flow velocity values at each measuring point The average flow velocity at all measuring points. For the velocity variance, For pressure measurement point index, This represents the total number of pressure measurement points. =15, For the first Pressure values at each measuring point This is the average pressure at all measuring points. For thick-walled hydrostatic pressure, For the coordination component index, The number of coordination dimensions. =5, For the first One coordination component.
[0048] The constraints of the multi-objective optimization framework include parameter range constraints, microstructure constraints, performance threshold constraints, and uncertainty coordination constraints, ensuring that the optimization is within the physical boundaries.
[0049] Parameter range constraints:
[0050] For example, =300°C, =500°C, , , =0.1, =1.0. Constraints are based on the physical limits of aluminum alloys, such as melting point and brittle transition temperature, to ensure parameters are within safe ranges.
[0051] Microstructure constraints:
[0052] in, Y represents the fraction of precipitation phase transformation. This is the lower limit of the grain size. =10μm, For reference strain rate, To avoid excessive refinement leading to brittleness. Precipitation phase transformation fraction. Satisfy 0≤ ≤1, and accumulates with temperature and deformation.
[0053] Performance threshold constraints:
[0054] Wherein, SDV represents the flow velocity deviation. =0.1m / s, SDP is the pressure deviation. =10MPa, HP is the hydrostatic pressure of the thick-walled structure. =100MPa, ensuring the performance threshold.
[0055] Uncertainty Coordination Constraints:
[0056] in, For the degree of uncertainty coordination, =0.5, For the velocity variance, For pressure variance, For the maximum permissible variance, =1, to ensure robustness.
[0057] The above constraints are applied as penalty functions in the NSGA-II iteration: if violated, Obj is increased with a penalty term.
[0058] Pareto's optimal solution set determination is based on the NSGA-II algorithm: after generating the solution set, the TOPSIS method is used for weighted evaluation, specifically, the weights are based on the importance of the objective, such as an intensity weight of 0.6 and an uncertainty weight of 0.4, and the optimal solution is selected by ranking. This determination incorporates a fuzzy rule base to ensure that the optimization direction is consistent with the degree of coordination.
[0059] Through iterative steps S1-S5 described above, this embodiment optimizes the hot deformation strength parameters of aluminum alloy. This embodiment can be implemented using a Python script: NSGA-II is performed using the pymoo library, fuzzy inference is handled using scikit-fuzzy methods, and iterative loops are run until Opt ≥ 0.8 or the maximum number of iterations is reached. Verification is achieved by comparing the predicted strength with experimental data using finite element simulation; the error is < 5%.
[0060] Example 2 This embodiment focuses on expanding the coordination degree solution in step S2 and the fuzzy inference engine training in step S3 to adapt to different aluminum alloy types. This embodiment emphasizes the dynamic adjustment of micro-parameters and incorporates elements of a three-dimensional thermal processing map to ensure the feasibility of the method in industrial scenarios.
[0061] In step S2, the extended coordination degree solution sub-step is performed: First, after extracting the parameters, the strain increment is calculated. .
[0062] Secondly, the dislocation density calculation uses an extended model: ,in, For Burgers vector, For recovery rate, .
[0063] Precipitation phase transformation , (Avrami index). Coordination , The dynamic weights adapt to the deformation stage: initial stage. =0.5, mid-term =0.4, end stage =0.1. The variants are simulated for microscopic evolution using MATLAB Simulink, ensuring a computation time of less than 1 second per iteration. In the solution... At that time, refer to the safe zone of the three-dimensional thermal processing diagram (temperature 200-500°C, strain rate 0.05-10). To optimize the compatibility of high-magnesium aluminum alloys (such as Al-4.0Mg), strain 0.3-1 is required to avoid unstable regions (such as adiabatic shear bands).
[0064] In step S3, the fuzzy inference engine constructs a variant: the input variables are expanded to... With uncertain inputs, such as noise δ=0.05, the fuzzy set is increased to 5 (extremely low, low, medium, high, extremely high), and the Gaussian parameter is initialized to μ=[0.1,0.3,0.5,0.7,0.9], σ=0.15. The fuzzy rule base is expanded to 25 rules, in Takagi-Sugeno form: rule example "If..." If the value is medium and δ is low, then the output is 0.6* +0.1*δ”. Training uses a hybrid algorithm: backpropagation updates the premise parameters, and least squares updates the conclusion parameters. There are 200 training datasets, 200 iterations, and an error threshold of 0.005. Validation: Input =0.72, δ=0.03, output μ=0.8.
[0065] Steps S4 and S5 are similar to those in Example 1, but the Pareto solution set size is increased to 200, incorporating Zener-Hollomon parameters. As an additional constraint (Z≤) During the iteration, if Not converged, adjust =1.2 Enhanced Penalty. Final Output Parameters: T=430°C , =0.9.
[0066] The variant was experimentally verified using plane strain thermal simulation (sample size 10×10×15mm, temperature 300-500°C, strain rate 0.05-10). The strain was reduced by 60% (the height was reduced), generating a three-dimensional hot working diagram (the safe zone avoids local deformation and dynamic strain failure), and the predicted strength error was <3%, proving that it is feasible. In high magnesium aluminum alloys, the optimized tensile strength increased from 322.3 MPa to 382.6 MPa, the elongation increased from 6.24% to 10.16%, and the metallographic structure showed dynamic recrystallization (no adiabatic shear band).
[0067] Example 3 This embodiment describes a system for optimizing the hot deformation strength parameters of aluminum alloys. The system includes an input module, a calculation module, an optimization module, and an output module. The system can be deployed on a computer, and module interaction can be achieved using Python or MATLAB.
[0068] Input module, to obtain initial parameters of thermal deformation of aluminum alloy; The calculation module solves for the thermal deformation strength compatibility of aluminum alloys based on the initial parameters of the aluminum alloy's hot deformation. The optimization module solves for the membership function based on the thermal deformation intensity coordination degree input to the fuzzy inference engine. The fuzzy inference engine adopts an adaptive neural fuzzy inference system, which maps the thermal deformation intensity coordination degree to membership values in the [0,1] interval and incorporates a multi-objective optimization framework to handle the uncertainty of thermal deformation. Based on the output value of the membership function, the optimization direction is determined by combining the fuzzy rule base and the Pareto optimal solution set. If the output value is lower than the dynamic threshold, the initial parameters of thermal deformation are adaptively adjusted to improve the thermal deformation intensity coordination degree. If the output value reaches or exceeds the dynamic threshold, the initial parameters of thermal deformation are not adjusted. The optimization calculation is iteratively executed until the optimization objective is reached, wherein the optimization objective is defined as maximizing the thermal deformation intensity coordination degree while minimizing the multi-objective optimization framework. The output module outputs robust thermal deformation strength parameters.
[0069] It is understood that the aluminum alloy hot deformation strength parameter optimization system provided in this embodiment and the aluminum alloy hot deformation strength parameter optimization method provided in the above embodiments are based on the same inventive concept. For more specific working principles of each module in the embodiments of the present invention, please refer to the above embodiments, and they will not be repeated in the embodiments of the present invention.
[0070] The above are merely preferred embodiments of the present invention and are not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
Claims
1. A method for optimizing the hot deformation strength parameters of aluminum alloys, characterized in that, The steps of this method include: Step S1: Obtain the initial parameters of the hot deformation of the aluminum alloy; Step S2: Solve for the thermal deformation strength compatibility of the aluminum alloy based on the initial parameters of the aluminum alloy's thermal deformation; Step S3: Based on the thermal deformation intensity coordination degree, input the fuzzy inference engine to solve the membership function. The fuzzy inference engine adopts an adaptive neural fuzzy inference system, which maps the thermal deformation intensity coordination degree to the membership value in the interval [0,1] and incorporates a multi-objective optimization framework to handle the uncertainty of thermal deformation. Step S4: Based on the output value of the membership function, the optimization direction is determined by combining the fuzzy rule base and the Pareto optimal solution set. If the output value is lower than the dynamic threshold, the initial parameters of thermal deformation are adaptively adjusted to improve the thermal deformation intensity coordination. If the output value reaches or exceeds the dynamic threshold, the initial parameters of thermal deformation are not adjusted. Step S5: Iteratively execute steps S2-S4 until the optimization objective is reached, obtain robust hot deformation strength parameters, and complete the optimization of aluminum alloy hot deformation strength parameters. The optimization objective is defined as maximizing the hot deformation strength coordination while minimizing the multi-objective optimization framework.
2. The method for optimizing the hot deformation strength parameters of aluminum alloys according to claim 1, characterized in that, The specific calculation steps for determining the compatibility of the hot deformation strength of the aluminum alloy are as follows: Temperature, strain rate, deformation, and initial dislocation density and initial grain size values from the initial parameters of aluminum alloy hot deformation are extracted as calculation inputs. The strain rate value is multiplied by the deformation value to obtain the strain increment. The current dislocation density value is obtained by balancing the dislocation hardening and recovery terms. The hardening term is a function of the strain increment, and the recovery term is a negative exponential function of the temperature value. The temperature and deformation values are substituted into an exponential function to calculate the conversion fraction of the precipitated phase from the initial state to the stable state. The conversion fraction accelerates with increasing temperature and accumulates with increasing deformation. The above results are aggregated by multidimensional fuzzy mapping. The dislocation density value, precipitation phase transformation fraction and initial grain size value are multiplied by dynamic weight coefficients based on deformation stage, and the weighted average is solved to obtain the thermal deformation strength compatibility.
3. The method for optimizing the hot deformation strength parameters of aluminum alloys according to claim 2, characterized in that, The specific construction steps of the fuzzy inference engine are as follows: Define input and output variables, take the thermal deformation strength compatibility as the input variable, and the membership value in the [0,1] interval as the output variable. Integrate the objective function in the multi-objective optimization framework as an auxiliary input to handle uncertainty; The input variables are fuzzified based on Gaussian membership functions, where the input variables correspond to the first fuzzy set, the second fuzzy set, and the third fuzzy set, respectively, and the parameters of the Gaussian membership functions are initialized. A fuzzy rule base is constructed based on the physical mechanism of hot deformation of aluminum alloy and historical hot deformation test data. The fuzzy rule base is configured in Takagi-Sugeno form, and the fuzzy rules cover all combinations of input variables. Historical aluminum alloy hot deformation test data were acquired and used as a training set. Gaussian membership function parameters and fuzzy rule weights were adaptively adjusted using backpropagation algorithm and least squares method. Verify whether the training error is less than the preset threshold. If not, iteratively adjust the Gaussian membership function parameters. If yes, complete the training of the fuzzy inference engine.
4. The method for optimizing the hot deformation strength parameters of aluminum alloys according to claim 3, characterized in that, The multi-objective optimization framework specifically aims to minimize the composite value of velocity exponential deviation, pressure logarithmic deviation, and thick-wall pressure exponential decay. Its objective function is calculated as follows: in, Let be the objective function. , , These are the adaptive weight coefficients, For flow velocity measurement point index, This represents the total number of flow velocity measurement points. For the first Flow velocity values at each measuring point The average flow velocity at all measuring points. For the velocity variance, For pressure measurement point index, This represents the total number of pressure measurement points. For the first Pressure values at each measuring point This is the average pressure at all measuring points. For thick-walled hydrostatic pressure, For the coordination component index, The number of coordination dimensions. For the first One coordination component.
5. The method for optimizing the hot deformation strength parameters of aluminum alloys according to claim 4, characterized in that, The constraints of the multi-objective optimization framework include parameter range constraints, microstructure constraints, performance threshold constraints, and uncertainty coordination constraints. The parameter range constraint is calculated using the following formula: in, , These are the lower and upper limits of the temperature, respectively. This is the heat distortion temperature. , These are the lower and upper limits of the strain rate, respectively. For strain rate, , These are the lower and upper limits of the deformation, respectively. This is the amount of deformation; The microstructure constraint is calculated using the following formula: in, For dislocation density, This is the upper limit of dislocation density. To restore the score, Grain size, This is the lower limit of the grain size. Used as a reference strain rate; The performance threshold constraint is calculated using the following formula: in, For the standard deviation of flow rate, This represents the upper limit of the standard deviation of the flow rate. For pressure standard deviation, This represents the upper limit of the pressure standard deviation. This is the lower limit of hydrostatic pressure for thick-walled structures. The calculation formula for the uncertainty coordination constraint is as follows: in, For the degree of uncertainty coordination, As the lower limit of the degree of coordination, For pressure variance, This represents the maximum permissible variance.
6. The method for optimizing the hot deformation strength parameters of aluminum alloys according to claim 5, characterized in that, The optimization direction is determined by combining the output value of the membership function with the fuzzy rule base and the Pareto optimal solution set, specifically as follows: Obtain the output value of the membership function; The output value is used as input, along with the initial parameters of the current hot deformation of the aluminum alloy and the microstructure vector, and then input into the fuzzy rule base. The preliminary optimization direction vector is generated through fuzzy rule base reasoning, and the constraints of the objective function of the multi-objective optimization framework are incorporated. The Pareto optimal solution set is generated based on the NSGA-II algorithm. The initial optimization direction vector is mapped onto the Pareto solution set, the weighted distance of each solution is calculated, and the nearest Pareto solution is selected. Determine the final optimization direction. If the selected Pareto solution satisfies the convergence condition, that is, the output value of the membership function increases to the set value, the objective function of the multi-objective optimization framework decreases to the set value, and all constraints are satisfied, then the adjusted parameter set is output. The parameter set includes the adjusted temperature, strain rate and deformation. Otherwise, the initial parameters of the thermal deformation are adaptively adjusted to improve the coordination of thermal deformation strength.
7. The method for optimizing the hot deformation strength parameters of aluminum alloys according to claim 6, characterized in that, The optimization objective is defined as maximizing the thermal deformation strength compatibility while minimizing the multi-objective optimization framework, and its specific calculation formula is as follows: in, To optimize the objective, For the thermal deformation strength compatibility, The penalty coefficient is... As an indicator of uncertainty, For indexing measurement points with uncertainties, The total number of measurement points with uncertainties. For the first Uncertainty rheological value at the measuring point For reference rheological values, For the first Stress value at measuring point For reference stress, For index, The normalization coefficient is... Index for the deformation stage, This represents the number of deformation stages. For the first Phase robust weights For the first The rate constant of the precipitation phase in stages, For the first Phase Zener-Hollomon parameters, For the first Stage index parameters, For the first Stage deformation amount For the first Stage strain rate, For the first Stage dislocation density, For the first Stage grain size, For maximum dislocation density, This represents the upper limit of the grain size. It is a micro-power exponent.
8. A system for optimizing the hot deformation strength parameters of aluminum alloys, characterized in that, include; Input module, to obtain initial parameters of thermal deformation of aluminum alloy; The calculation module solves for the thermal deformation strength compatibility of aluminum alloys based on the initial parameters of the aluminum alloy's hot deformation. The optimization module solves for the membership function based on the thermal deformation intensity coordination degree input to the fuzzy inference engine. The fuzzy inference engine adopts an adaptive neural fuzzy inference system, which maps the thermal deformation intensity coordination degree to membership values in the [0,1] interval and incorporates a multi-objective optimization framework to handle the uncertainty of thermal deformation. Based on the output value of the membership function, the optimization direction is determined by combining the fuzzy rule base and the Pareto optimal solution set. If the output value is lower than the dynamic threshold, the initial parameters of thermal deformation are adaptively adjusted to improve the thermal deformation intensity coordination degree. If the output value reaches or exceeds the dynamic threshold, the initial parameters of thermal deformation are not adjusted. The optimization calculation is iteratively executed until the optimization objective is reached, wherein the optimization objective is defined as maximizing the thermal deformation intensity coordination degree while minimizing the multi-objective optimization framework. The output module outputs robust thermal deformation strength parameters.
9. The aluminum alloy hot deformation strength parameter optimization system according to claim 8, characterized in that, The calculation module further includes a coordination degree solution submodule, which extracts temperature, strain rate, deformation amount, and initial dislocation density and initial grain size values from the initial parameters of aluminum alloy hot deformation as calculation inputs; multiplies the strain rate value by the deformation amount to obtain the strain increment; balances the dislocation hardening and recovery terms to obtain the current dislocation density value, where the hardening term is a function of the strain increment and the recovery term is a negative exponential function of the temperature value; substitutes the temperature value and deformation amount into the exponential function to calculate the transformation fraction of the precipitated phase from the initial state to the stable state, the transformation fraction accelerates with increasing temperature and accumulates with increasing deformation amount; aggregates the above results through multidimensional fuzzy mapping, multiplies the dislocation density value, precipitated phase transformation fraction, and initial grain size value by dynamic weighting coefficients based on the deformation stage, and solves the weighted average to obtain the hot deformation strength coordination degree.
10. The aluminum alloy hot deformation strength parameter optimization system according to claim 9, characterized in that, The output module further includes a visualization submodule for performing visualization of the output thermal deformation strength parameters.