Micro spring washer production method based on genetic algorithm optimization

Abstract

The invention discloses a micro spring washer production method based on genetic algorithm optimization, and the method comprises the following steps: S1, initializing a technological parameter set, and generating a preliminary optimization scheme through a morphological optimization control model; s2, mapping high-dimensional parameters to a low-dimensional morphological feature space through self-organizing mapping, and implementing adaptive variation adjustment; s3, integrating the processed morphological characteristics and technological parameters into chromosome codes, and performing multi-generation iterative optimization; s4, combining an evolutionary game model and fuzzy logic control, and dynamically adjusting a process parameter combination; s5, cooperatively regulating and controlling technological parameters through crossover and mutation operations; s6, performing fitness evaluation, and adjusting genetic algorithm parameters; and S7, after a convergence condition is reached, outputting an optimal process parameter, and applying the optimal process parameter to actual production. The method can significantly improve the production efficiency and the product quality, and is suitable for the precision manufacturing field of complex process optimization.

Description

Technical Field

[0001] The invention relates to the technical field of genetic algorithms, and in particular to a method for producing micro spring washers based on genetic algorithm optimization. Background Art

[0002] First of all, the manufacturing technology of micro-mechanical components occupies an extremely important position in modern industry, especially in the fields of electronic equipment, precision instruments, and aerospace. As a key micro-mechanical component, the accuracy and stability of the production process of micro-spring washers directly affect the performance and service life of the product. However, the traditional production method of micro-spring washers mainly relies on empirical design and manual adjustment of process parameters, which has problems such as low efficiency, large quality fluctuations, and high production costs. In these methods, process parameters such as pressure, temperature, cutting speed, and material properties are usually determined by trial and error or empirical formulas, which makes it difficult to optimize multiple parameters at the same time and ensure product consistency. In addition, with the increasing complexity of product design, traditional methods are unable to cope with multi-objective optimization problems.

[0003] Secondly, existing process optimization methods are usually based on simple algorithms or heuristic methods, such as linear regression, stepwise optimization or a single application of genetic algorithms. Although these methods perform well in certain single-objective optimizations, they have limited effectiveness in complex process environments involving multiple objectives and multiple constraints. For example, although a single genetic algorithm can optimize process parameters to a certain extent, it is prone to fall into local optimal solutions due to its limitations in population diversity and adaptability, resulting in unsatisfactory accuracy and consistency of the produced micro spring washers. In addition, traditional methods often lack the ability to adjust in real time when dealing with dynamic production environments, and cannot effectively cope with fluctuations and changes in the production process.

[0004] Third, although some studies have attempted to improve process optimization by introducing some advanced algorithms such as neural networks and fuzzy logic control, the application of these methods in the field of micro-machine manufacturing still faces challenges. Specifically, it is difficult for existing technologies to simultaneously consider the coupling relationship between microstructure and macro process parameters, and there is a lack of methods that can accurately map low-dimensional morphological features in high-dimensional parameter space. In addition, existing optimization methods often have difficulty in handling the complex nonlinear relationship between process parameters and morphological features, resulting in insufficient data fitting capabilities during the optimization process, which ultimately affects the product quality of micro spring washers.

[0005] In summary, the main defects of the existing technology in the production of micro spring washers include: 1. Lack of global optimization capabilities under multi-objective and multi-constraint conditions during process parameter optimization; 2. Difficulty in handling real-time adjustment requirements in a dynamic production environment, resulting in instability in the production process; 3. When dealing with the coupling relationship between process parameters and morphological characteristics, there is a lack of effective methods for accurate low-dimensional mapping, resulting in insufficient optimization accuracy; 4. Lack of adaptive control of process parameter variations can easily lead to the optimization process converging to a local optimal solution prematurely. Therefore, how to provide a micro spring washer production method based on genetic algorithm optimization is an urgent problem that technicians in this field need to solve. Summary of the invention

[0006] One purpose of the present invention is to propose a method for producing micro spring washers based on genetic algorithm optimization. The present invention combines genetic algorithm, evolutionary game model and fuzzy logic control mechanism, and realizes the precise control of process parameters and the best matching of morphological characteristics in the production process of micro spring washers through global optimization and dynamic adjustment of process parameters. The method effectively improves production efficiency, product quality and process stability, and is suitable for complex process optimization under multi-objective and multi-constraint conditions.

[0007] A method for producing a micro spring gasket based on genetic algorithm optimization according to an embodiment of the present invention comprises the following steps:

[0008] S1. Initialize the process parameter set, including pressure, temperature, cutting speed and material microstructure parameters, use the morphological optimization control model to analyze the process parameters, and generate a preliminary optimization plan;

[0009] S2, using self-organizing mapping to map the results of morphological optimization from high-dimensional parameter space to low-dimensional morphological feature space, and adjusting the process parameters through adaptive mutation strategy;

[0010] S3, integrating the morphological features and process parameters after self-organizing mapping into the chromosome encoding of the genetic algorithm, performing multiple generations of iterations, and optimizing the matching between microstructure and macro process parameters;

[0011] S4. In the process of genetic algorithm evolution, combined with the evolutionary game model, the process parameter combination is dynamically adjusted through competition and cooperation of production strategies, and the evolutionary strategy is adjusted according to real-time production environment data through the fuzzy logic control mechanism;

[0012] S5. In crossover and mutation operations, evolutionary game and fuzzy logic control are used to coordinately regulate key process parameters, so that the optimization results can maintain stability and consistency under complex conditions;

[0013] S6. After each iteration, the morphological optimization and self-organizing map model are used to evaluate the fitness, and the genetic algorithm parameters are adjusted by combining evolutionary game and fuzzy logic control;

[0014] S7. After reaching the convergence condition or the process target reaches the predetermined standard, the individual with the highest fitness is output as the final production process parameter and applied to actual production.

[0015] Optionally, the S1 includes the following steps:

[0016] S11, set the initial process parameter set X = {P, T, v, M, σ, ∈, ξ}, where P is the pressure parameter, T is the temperature parameter, v is the cutting speed parameter, M is the material microstructure parameter, σ is the yield strength, ∈ is the strain characteristic, and ξ is the morphological adaptability factor, which is used to measure the influence of the process parameters on the microstructure morphology;

[0017] S12, establish a morphology-process coupling optimization model, couple the microstructure morphology parameters with the process parameter set X, and calculate the morphology adaptability factor ξ, which is used to represent the comprehensive influence of different process parameters on the microstructure morphology;

[0018] S13, according to the morphology-process coupling optimization model, the optimization target value ξ0 is set, and through iterative calculation, each parameter in the process parameter set X is adjusted to achieve the best adaptability with the target value ξ0, so as to optimize the microstructure morphology characteristics;

[0019] S14, performing sensitivity analysis on the optimized process parameter set X, introducing a disturbance factor δξ according to the change of the morphological adaptability factor ξ, evaluating its influence on the microstructure morphology by adjusting the process parameter set, and performing parameter correction;

[0020] S15, repeat the optimization and sensitivity analysis steps until the morphological adaptability factor ξ reaches the best match with the target value ξ0, and finally generate the optimized process parameter set X'', which is used as the preliminary optimization scheme and input into the subsequent optimization steps of the genetic algorithm.

[0021] Optionally, S2 includes the following steps:

[0022] S21, take the morphologically optimized process parameter set X' as input, construct a self-organizing mapping network, define the input layer of the network as the high-dimensional process parameter X', and the output layer as the low-dimensional morphological feature space Z = z1, z2, ..., z n}, where each eigenvalue z in Z i Represents the projection of morphological features in low-dimensional space;

[0023] S22. Based on the nonlinear coupling relationship between the microstructure of the spring gasket material and the process parameters, a nonlinear mapping formula of the self-organizing map is constructed to map the process parameters X' to the low-dimensional morphological feature space Z:

[0024]

[0025] Among them, w ij is the weight of the mapping network, α j and β j is the nonlinear coefficient of the control process parameters, γ j is the morphological adaptability factor coefficient, δ j is the nonlinear oscillation adjustment factor, λ j is the frequency control factor, is the derivative of the morphological adaptability factor with respect to the process parameters;

[0026] S23, by training the self-organizing map network, optimize the network weight W and control coefficient α j , β j , γ j ,δ j and λ j , minimize the error function ε:

[0027]

[0028] in, is the expected morphological feature value, λ1 is the sensitivity matching factor, and λ2 is the weight regularization factor used to prevent overfitting;

[0029] S24. Aiming at the nonlinear behavior in the mapping process, a multi-scale morphological feature adjustment strategy is used to grade the main and secondary morphological features;

[0030] S25. Combining the morphological feature gain adjustment and the process parameter sensitivity, each parameter in the process parameter set X" is mutated and adjusted by adaptive mutation:

[0031]

[0032] Among them, κ i is to adjust the morphological feature gain coefficient, η is the mutation rate control factor;

[0033] S26, iteratively adjusting and optimizing the process parameter set X'' until the optimization target is met, and finally taking the adjusted process parameter set X''' as the final optimization result and inputting it into the subsequent steps of the genetic algorithm.

[0034] Optionally, S3 includes the following steps:

[0035] S31, the morphological feature set Z after self-organizing mapping processing is n} and the process parameter set X'' are integrated into the chromosome encoding C = {Z, X''} of the genetic algorithm;

[0036] S32, initialize the population P = {C1, C2, ..., C k}, where each individual C i Represents different combinations of morphological characteristics and process parameters; the diversity of the population is achieved by perturbing the morphological characteristics and process parameters, so that the initial population has a wide distribution between the morphological characteristics and process parameters;

[0037] S33. Define fitness function To evaluate the quality of each chromosome; the fitness function is formulated according to the comprehensive optimization goal of morphological characteristics and process parameters:

[0038]

[0039] Among them, ω1 and ω2 are weight factors, α i and β j is the nonlinear adjustment coefficient, and are the desired morphological characteristic values ​​and process parameter values, respectively;

[0040] S34. In the crossover operation of the genetic algorithm, a hierarchical crossover strategy is adopted to perform a crossover operation on the morphological feature subset Z and a crossover operation on the process parameter subset X'':

[0041] C new =α·C parent1 +(1-α)·C parent2 +γ·sin(π·(C parent1 -C parent2 ));

[0042] Among them, C new is the newly generated offspring chromosome, α is the crossover rate control factor, γ is the nonlinear crossover adjustment factor, sin(π·(C parent1 -C parent2 )) is the introduced nonlinear interference term;

[0043] S35. In the mutation operation, adaptive mutation strategies are respectively implemented for the morphological feature subset Z and the process parameter subset X'' in the chromosome, and the mutation rate η is dynamically adjusted according to the fitness distribution of the current population;

[0044] S36. After multiple generations of iterations, monitor the fitness function The chromosome with the highest fitness is selected to enter the next generation until the fitness function converges or reaches the preset optimization goal. The final output chromosome C opt Including the optimized morphological features Z opt and process parameters X”’ as the optimal process configuration for the production of micro spring washers.

[0045] Optionally, S4 includes the following steps:

[0046] S41. In the evolution process of genetic algorithm, an evolutionary game model is constructed, each chromosome C=Z,X”'} is regarded as a game body, and its strategy space is set to S={s1,s2,...,s m}, where each strategy s i Corresponding to a process parameter adjustment plan;

[0047] S42. Define the payoff function U(C,s i ), the payment function is based on the optimization objectives of the process parameters and the adaptability of the morphological characteristics:

[0048]

[0049] Among them, ω3 and ω4 are weight factors, λ is the adjustment coefficient, is the fitness function of the chromosome, and the payment function is used to evaluate the specific strategy s i The optimization effect under

[0050] S43. In each generation of evolution, the game body selects strategies through the evolutionary game model to simulate the process of competition and cooperation; the strategy selection adopts a mixed strategy game model, in which each chromosome adjusts its process parameter combination according to the game results:

[0051]

[0052] Where η is the learning rate is the partial derivative of the payment function with respect to the strategy. After the strategy is updated, the process parameter combination of chromosome C is adjusted;

[0053] S44, combined with the fuzzy logic control mechanism, the strategy space S of the evolutionary game is dynamically adjusted through the input of real-time production environment data;

[0054] S45, in the process of evolution, chromosomes are adjusted through the payment function U(C,s i ) and fitness function The combined effect of the two factors is to continuously optimize the process parameter combination and select the strategy combination with the highest fitness to enter the next generation of iteration;

[0055] S46, through multiple iterations of evolutionary game and fuzzy logic control mechanism, the optimized chromosome C is generated opt , C opt Contains the optimal morphological feature Z opt and process parameter combination X”” and serve as the optimal configuration solution for subsequent production processes.

[0056] Optionally, S5 includes the following steps:

[0057] S51. In the crossover operation of the genetic algorithm, for the morphological feature subset of chromosomes Z = z1, z2, ..., z n}Perform multi-point crossover and form a new chromosome Z by setting the crossover points new ;

[0058] S52, in the crossover operation of the process parameter subset X'', the crossover rate is dynamically adjusted by fuzzy control, and the crossover operation is adjusted according to the real-time production environment;

[0059] S53. After completing the crossover operation, the process parameters are further optimized using the evolutionary game model; each chromosome is regarded as a game subject, and the optimal combination of process parameters is achieved by adjusting the strategy space. The strategy update formula is:

[0060]

[0061] Where η is the learning rate factor, is the sensitivity of process parameters to strategy, cos(π·s j ) is the periodic adjustment factor;

[0062] S54, when performing mutation operation, a multi-level dynamic mutation strategy is adopted to perform adaptive mutation on the morphological feature Z and process parameter X'' in the chromosome respectively:

[0063]

[0064] Among them, γ i is the variation enhancement coefficient of morphological characteristics, β i is the morphological feature variation suppression coefficient, λ j is the process parameter variation enhancement coefficient, α j is the process parameter variation suppression coefficient, and by adjusting these parameters, the direction and amplitude of variation can be controlled;

[0065] S55. In each iteration, the mutation rate and mutation direction are adjusted according to the real-time production environment data through the fuzzy logic control mechanism;

[0066] S56, the results of comprehensive evolutionary game and fuzzy logic control, combined with fitness function and the payment function U(C,si ), evaluate the newly generated chromosomes, and retain the chromosomes with the highest fitness and stability to enter the next generation;

[0067] S57, through multiple rounds of iterative optimization, the optimal chromosome C is finally generated opt , C opt Chromosomes contain morphological features that exhibit optimal stability and consistency under complex production conditions. opt and process parameter combination X””;

[0068] Optionally, the S6 includes the following steps:

[0069] S61. After each iteration, the fitness of all chromosomes C = {Z, X'} is evaluated. The fitness evaluation is based on the joint feedback of the morphological optimization model and the self-organizing map type; the fitness function Combining the coupling relationship between morphological characteristics and process parameters, the nonlinear weighted formula is used to calculate:

[0070]

[0071] Among them, ω1 and ω2 are the weight coefficients of morphological characteristics and process parameters, α i and β j is the nonlinear adjustment factor, and are the target morphological characteristic values ​​and process parameter values;

[0072] S62. Based on the fitness evaluation results, the key parameters of the genetic algorithm are dynamically adjusted; the adjustment formula of the crossover rate α is based on the interactive influence of morphological characteristics and process parameters, as follows:

[0073]

[0074] Among them, γ is the adjustment coefficient, θ is the phase adjustment factor, is the change in fitness function;

[0075] S63, the mutation rate η is adjusted in real time, combined with the fuzzy logic control mechanism, and the frequency and intensity of the mutation operation are adjusted through the feedback of environmental data. The adjustment formula of the mutation rate is:

[0076]

[0077] Among them, μ is the fuzzy control gain coefficient, λ is the adjustment factor, fuzz(E) is the fuzzification processing result of environmental data, is the fitness change;

[0078] S64. In each iteration, the competition and cooperation between different chromosomes are simulated through the evolutionary game model, and the strategy of dynamically adjusting the process parameters is:

[0079]

[0080] Among them, ζ is the strategy adjustment coefficient, κ i It is the strategy adjustment factor. Through the combination of sine and hyperbolic cosine functions, the strategy adjustment can better adapt to the complex nonlinear relationship between process parameters.

[0081] S65. According to the feedback from the morphological optimization model and the self-organizing map model, chromosomes with low fitness are eliminated, and at the same time, some chromosomes are re-mutated in combination with the fuzzy logic control mechanism to maintain the diversity of the population and the global search capability;

[0082] S66, after multiple generations of iterations, the chromosome C with the highest fitness and the best stability was finally selected opt , which contains the optimized morphological features Z opt and process parameter combination X”” to provide the best production process configuration solution;

[0083] S67, the optimal chromosome C opt Applied to the actual production process, and continuously monitor and adjust the optimization parameters during the production process.

[0084] The beneficial effects of the present invention are:

[0085] (1) The present invention proposes a method for producing micro spring washers based on genetic algorithm optimization. By initializing, optimizing and dynamically adjusting process parameters, combined with morphological optimization control, self-organizing mapping and evolutionary game model, the production efficiency and product quality of micro spring washers are significantly improved. The present invention uses a morphological optimization control model to conduct a preliminary analysis of process parameters and generate an optimization scheme, which effectively reduces production errors caused by improper parameter settings and ensures the best match between microstructure and macro process parameters, especially in the field of precision manufacturing.

[0086] (2) The present invention introduces a self-organizing mapping model and an evolutionary game model, maps high-dimensional process parameters to a low-dimensional morphological feature space, and combines the evolutionary game mechanism to achieve dynamic adjustment of parameters. This method effectively solves the limitations of traditional single optimization algorithms under multi-objective and multi-constrained process conditions, enables real-time optimization of process parameters in a complex production environment, ensures the stability and adaptability of process parameters during the production process, and significantly improves the consistency and reliability of products.

[0087] (3) The present invention adaptively adjusts the process parameters by combining the fuzzy logic control mechanism in the evolution process of the genetic algorithm. The real-time feedback mechanism of the fuzzy logic control is used to dynamically adjust the optimization strategy according to the changes in the production environment, thereby avoiding the process disorder caused by environmental fluctuations in the traditional method, effectively improving the stability of the production process and the accuracy of the final product, and ensuring the high-quality output of the micro spring gasket under different process conditions.

[0088] (4) The present invention uses multiple rounds of iterative optimization to evaluate the fitness of chromosomes after each iteration, and combines morphological optimization with the self-organizing map model to adjust the process parameters, gradually optimize the genetic algorithm parameters, and finally reach the convergence condition or process target standard. This process makes the optimization of process parameters more efficient and accurate under a multi-objective and multi-constraint environment, significantly improving the production efficiency and product qualification rate of micro spring washers. BRIEF DESCRIPTION OF THE DRAWINGS

[0089] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0090] Figure 1 A flowchart of a method for producing micro spring washers based on genetic algorithm optimization proposed by the present invention;

[0091] Figure 2 A flow chart of the collaborative optimization of process parameters using the evolutionary game model and fuzzy logic control mechanism proposed in the present invention;

[0092] Figure 3 This is a schematic diagram of process parameter optimization and fitness evaluation during the multi-generation iteration process of the genetic algorithm proposed in the present invention. DETAILED DESCRIPTION

[0093] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, which only illustrate the basic structure of the present invention in a schematic manner, and therefore only show the components related to the present invention.

[0094] refer to Figure 1-3 , a method for producing a micro spring gasket based on genetic algorithm optimization, comprising the following steps:

[0095] S1. Initialize the process parameter set, including pressure, temperature, cutting speed and material microstructure parameters, use the morphological optimization control model to analyze the process parameters, and generate a preliminary optimization plan;

[0096] S2, using self-organizing mapping to map the results of morphological optimization from high-dimensional parameter space to low-dimensional morphological feature space, and adjusting the process parameters through adaptive mutation strategy;

[0097] S3, integrating the morphological features and process parameters after self-organizing mapping into the chromosome encoding of the genetic algorithm, performing multiple generations of iterations, and optimizing the matching between microstructure and macro process parameters;

[0098] S4. In the process of genetic algorithm evolution, combined with the evolutionary game model, the process parameter combination is dynamically adjusted through competition and cooperation of production strategies, and the evolutionary strategy is adjusted according to real-time production environment data through the fuzzy logic control mechanism;

[0099] S5. In crossover and mutation operations, evolutionary game and fuzzy logic control are used to coordinately regulate key process parameters, so that the optimization results can maintain stability and consistency under complex conditions;

[0100] S6. After each iteration, the morphological optimization and self-organizing map model are used to evaluate the fitness, and the genetic algorithm parameters are adjusted by combining evolutionary game and fuzzy logic control;

[0101] S7. After reaching the convergence condition or the process target reaches the predetermined standard, the individual with the highest fitness is output as the final production process parameter and applied to actual production.

[0102] In this implementation, S1 includes the following steps:

[0103] S11, set the initial process parameter set X = {P, T, v, M, σ, ∈, ξ}, where P is the pressure parameter, T is the temperature parameter, v is the cutting speed parameter, M is the material microstructure parameter, σ is the yield strength, ∈ is the strain characteristic, and ξ is the morphological adaptability factor, which is used to measure the influence of the process parameters on the microstructure morphology;

[0104] S12, establish a morphology-process coupling optimization model, couple the microstructure morphology parameters with the process parameter set X, and calculate the morphology adaptability factor ξ, which is used to represent the comprehensive influence of different process parameters on the microstructure morphology;

[0105] S13, according to the morphology-process coupling optimization model, the optimization target value ξ0 is set, and through iterative calculation, each parameter in the process parameter set X is adjusted to achieve the best adaptability with the target value ξ0, so as to optimize the microstructure morphology characteristics;

[0106] S14, performing sensitivity analysis on the optimized process parameter set X, introducing a disturbance factor δξ according to the change of the morphological adaptability factor ξ, evaluating its influence on the microstructure morphology by adjusting the process parameter set, and performing parameter correction;

[0107] S15, repeat the optimization and sensitivity analysis steps until the morphological adaptability factor ξ reaches the best match with the target value ξ0, and finally generate the optimized process parameter set X'', which is used as the preliminary optimization scheme and input into the subsequent optimization steps of the genetic algorithm.

[0108] In this implementation, S2 includes the following steps:

[0109] S21, take the morphologically optimized process parameter set X' as input, construct a self-organizing mapping network, define the input layer of the network as the high-dimensional process parameter X', and the output layer as the low-dimensional morphological feature space Z = z1, z2, ..., z n}, where each eigenvalue z in Z i Represents the projection of morphological features in low-dimensional space;

[0110] S22. Based on the nonlinear coupling relationship between the microstructure of the spring gasket material and the process parameters, a nonlinear mapping formula of the self-organizing map is constructed to map the process parameters X' to the low-dimensional morphological feature space Z:

[0111]

[0112] Among them, w ij is the weight of the mapping network, α j and β j is the nonlinear coefficient of the control process parameters, γ j is the morphological adaptability factor coefficient, δ j is the nonlinear oscillation adjustment factor, λ j is the frequency control factor, is the derivative of the morphological adaptability factor with respect to the process parameters;

[0113] S23, by training the self-organizing map network, optimize the network weight W and control coefficient α j , β j , γ j ,δ j and λ j , minimize the error function v:

[0114]

[0115] in, is the expected morphological feature value, λ1 is the sensitivity matching factor, and λ2 is the weight regularization factor used to prevent overfitting;

[0116] S24. Aiming at the nonlinear behavior in the mapping process, a multi-scale morphological feature adjustment strategy is used to grade the main and secondary morphological features;

[0117] S25. Combining the morphological feature gain adjustment and the process parameter sensitivity, each parameter in the process parameter set X" is mutated and adjusted by adaptive mutation:

[0118]

[0119] Among them, κ i is to adjust the morphological feature gain coefficient, η is the mutation rate control factor;

[0120] S26, iteratively adjusting and optimizing the process parameter set X'' until the optimization target is met, and finally taking the adjusted process parameter set X''' as the final optimization result and inputting it into the subsequent steps of the genetic algorithm.

[0121] In this implementation, S3 includes the following steps:

[0122] S31, the morphological feature set Z after self-organizing mapping processing is n} and the process parameter set X'' are integrated into the chromosome encoding C = {Z, X''} of the genetic algorithm;

[0123] S32, initialize the population P = {C1, C2, ..., C k}, where each individual C i Represents different combinations of morphological characteristics and process parameters; the diversity of the population is achieved by perturbing the morphological characteristics and process parameters, so that the initial population has a wide distribution between the morphological characteristics and process parameters;

[0124] S33. Define fitness function To evaluate the quality of each chromosome; the fitness function is formulated according to the comprehensive optimization goal of morphological characteristics and process parameters:

[0125]

[0126] Among them, ω1 and ω2 are weight factors, α i and β j is the nonlinear adjustment coefficient, and are the desired morphological characteristic values ​​and process parameter values, respectively;

[0127] S34. In the crossover operation of the genetic algorithm, a hierarchical crossover strategy is adopted to perform a crossover operation on the morphological feature subset Z and a crossover operation on the process parameter subset X'':

[0128] C new =α·C parent1 +(1-α)·C parent2 +γ·sin(π·(C parent1 -C parent2 ));

[0129] Among them, C new is the newly generated offspring chromosome, α is the crossover rate control factor, γ is the nonlinear crossover adjustment factor, sin(π·(C parent1 -C parent2 )) is the introduced nonlinear interference term;

[0130] S35. In the mutation operation, adaptive mutation strategies are respectively implemented for the morphological feature subset Z and the process parameter subset X'' in the chromosome, and the mutation rate η is dynamically adjusted according to the fitness distribution of the current population;

[0131] S36. After multiple generations of iterations, monitor the fitness function The chromosome with the highest fitness is selected to enter the next generation until the fitness function converges or reaches the preset optimization goal. The final output chromosome C opt Including the optimized morphological features Z opt and process parameters X”’ as the optimal process configuration for the production of micro spring washers.

[0132] In this implementation, S4 includes the following steps:

[0133] S41. In the evolution process of genetic algorithm, an evolutionary game model is constructed, each chromosome C=Z,X”'} is regarded as a game body, and its strategy space is set to S={s1,s2,...,s m}, where each strategy s i Corresponding to a process parameter adjustment plan;

[0134] S42. Define the payoff function U(C,s i ), the payment function is based on the optimization objectives of the process parameters and the adaptability of the morphological characteristics:

[0135]

[0136] Among them, ω3 and ω4 are weight factors, λ is the adjustment coefficient, is the fitness function of the chromosome, and the payment function is used to evaluate the specific strategy s i The optimization effect under

[0137] S43. In each generation of evolution, the game body selects strategies through the evolutionary game model to simulate the process of competition and cooperation; the strategy selection adopts a mixed strategy game model, in which each chromosome adjusts its process parameter combination according to the game results:

[0138]

[0139] Where η is the learning rate is the partial derivative of the payment function with respect to the strategy. After the strategy is updated, the process parameter combination of chromosome C is adjusted;

[0140] S44, combined with the fuzzy logic control mechanism, the strategy space S of the evolutionary game is dynamically adjusted through the input of real-time production environment data;

[0141] S45, in the process of evolution, chromosomes are adjusted through the payment function U(C,s i ) and fitness function The combined effect of the two factors is to continuously optimize the process parameter combination and select the strategy combination with the highest fitness to enter the next generation of iteration;

[0142] S46, through multiple iterations of evolutionary game and fuzzy logic control mechanism, the optimized chromosome C is generated opt , C opt Contains the optimal morphological feature Z opt and process parameter combination X”” and serve as the optimal configuration solution for subsequent production processes.

[0143] In this implementation, S5 includes the following steps:

[0144] S51. In the crossover operation of the genetic algorithm, for the morphological feature subset of chromosomes Z = z1, z2, ..., z n}Perform multi-point crossover and form a new chromosome Z by setting the crossover points new ;

[0145] S52, in the crossover operation of the process parameter subset X'', the crossover rate is dynamically adjusted by fuzzy control, and the crossover operation is adjusted according to the real-time production environment;

[0146] S53. After completing the crossover operation, the process parameters are further optimized using the evolutionary game model; each chromosome is regarded as a game subject, and the optimal combination of process parameters is achieved by adjusting the strategy space. The strategy update formula is:

[0147]

[0148] Where η is the learning rate factor, is the sensitivity of process parameters to strategy, cos(π·s j ) is the periodic adjustment factor;

[0149] S54, when performing mutation operation, a multi-level dynamic mutation strategy is adopted to perform adaptive mutation on the morphological feature Z and process parameter X'' in the chromosome respectively:

[0150]

[0151] Among them, γi is the variation enhancement coefficient of morphological characteristics, β i is the morphological feature variation suppression coefficient, λ j is the process parameter variation enhancement coefficient, α j is the process parameter variation suppression coefficient, and by adjusting these parameters, the direction and amplitude of variation can be controlled;

[0152] S55. In each iteration, the mutation rate and mutation direction are adjusted according to the real-time production environment data through the fuzzy logic control mechanism;

[0153] S56, the results of comprehensive evolutionary game and fuzzy logic control, combined with fitness function and the payment function U(C,s i ), evaluate the newly generated chromosomes, and retain the chromosomes with the highest fitness and stability to enter the next generation;

[0154] S57, through multiple rounds of iterative optimization, the optimal chromosome C is finally generated opt , C opt Chromosomes contain morphological features that exhibit optimal stability and consistency under complex production conditions. opt and process parameter combination X””;

[0155] In this implementation, S6 includes the following steps:

[0156] S61. After each iteration, the fitness of all chromosomes C = {Z, X'} is evaluated. The fitness evaluation is based on the joint feedback of the morphological optimization model and the self-organizing map type; the fitness function Combining the coupling relationship between morphological characteristics and process parameters, the nonlinear weighted formula is used to calculate:

[0157]

[0158] Among them, ω1 and ω2 are the weight coefficients of morphological characteristics and process parameters, α i and β j is the nonlinear adjustment factor, and are the target morphological characteristic values ​​and process parameter values;

[0159] S62. Based on the fitness evaluation results, the key parameters of the genetic algorithm are dynamically adjusted; the adjustment formula of the crossover rate α is based on the interactive influence of morphological characteristics and process parameters, as follows:

[0160]

[0161] Among them, γ is the adjustment coefficient, θ is the phase adjustment factor, is the change in fitness function;

[0162] S63, the mutation rate η is adjusted in real time, combined with the fuzzy logic control mechanism, and the frequency and intensity of the mutation operation are adjusted through the feedback of environmental data. The adjustment formula of the mutation rate is:

[0163]

[0164] Among them, μ is the fuzzy control gain coefficient, λ is the adjustment factor, fuzz(E) is the fuzzification processing result of environmental data, is the fitness change;

[0165] S64. In each iteration, the competition and cooperation between different chromosomes are simulated through the evolutionary game model, and the strategy of dynamically adjusting the process parameters is:

[0166]

[0167] Among them, ζ is the strategy adjustment coefficient, κ i It is the strategy adjustment factor. Through the combination of sine and hyperbolic cosine functions, the strategy adjustment can better adapt to the complex nonlinear relationship between process parameters.

[0168] S65. According to the feedback from the morphological optimization model and the self-organizing map model, chromosomes with low fitness are eliminated, and at the same time, some chromosomes are re-mutated in combination with the fuzzy logic control mechanism to maintain the diversity of the population and the global search capability;

[0169] S66, after multiple generations of iterations, the chromosome C with the highest fitness and the best stability was finally selected opt , which contains the optimized morphological features Z opt and process parameter combination X”” to provide the best production process configuration solution;

[0170] S67, the optimal chromosome C opt Applied to the actual production process, and continuously monitor and adjust the optimization parameters during the production process.

[0171] Embodiment 1:

[0172] In a well-known electronic equipment manufacturer, the micro spring washers produced by the company are mainly used in high-end smartphones, wearable devices and precision medical instruments. The production of these washers requires extremely high precision and consistency to ensure their stability and reliability in use. However, since traditional production methods mainly rely on manual setting of process parameters and a single optimization algorithm, the product quality is unstable during the production process, the defective rate is high, and it is difficult to improve production efficiency. In the context of increasingly fierce market competition, the company urgently needs an advanced production method that can achieve global optimization of process parameters under multi-objective and multi-constraint conditions, thereby improving product quality and reducing production costs.

[0173] In order to solve the above problems, the company introduced the micro spring gasket production method based on genetic algorithm optimization proposed in this invention. In actual production, engineers first use the morphological optimization control model to initialize the analysis of key process parameters (including pressure, temperature, cutting speed and material microstructure parameters) and generate a preliminary optimization plan. This step significantly reduces the time of trial and error based on experience in the past and improves the efficiency of production preparation.

[0174] After obtaining the preliminary optimization plan, engineers used the self-organizing mapping model to map the high-dimensional process parameters after morphological optimization to the low-dimensional morphological feature space, and further optimized these parameters through the adaptive mutation strategy. Unlike traditional manual adjustments, this process realizes real-time adjustments for different batches of production, making the micro spring washers produced more consistent and stable in various indicators.

[0175] In order to ensure the best optimization of process parameters in the production process, the company further introduced evolutionary game models and fuzzy logic control mechanisms. The evolutionary game model simulates the competition and cooperation between different process parameter combinations, continuously adjusts the parameter combination in multiple iterations, and finally achieves the global optimal solution. At the same time, the fuzzy logic control mechanism dynamically adjusts the process parameters according to the changes in the real-time production environment, ensuring that the process parameters can respond to changes in time under different production conditions, avoiding process disorders caused by environmental fluctuations.

[0176] Data Proof

[0177] The company applied the method of the present invention in actual production and conducted a one-year experiment. The results showed that production efficiency and product quality were significantly improved. The following is a detailed comparison of the actual production data before and after the experiment:

[0178] Table 1 Data table

[0179]

[0180]

[0181] It can be seen from the data that the method of the present invention significantly improves production efficiency and product quality in actual production, reduces the defective rate, optimizes material utilization, and ensures product consistency and process stability. By applying the method of the present invention, the enterprise not only effectively solves a number of technical problems in previous production, but also significantly improves the economic benefits of the enterprise, which wins it a greater advantage in market competition.

[0182] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A method for producing micro spring washers based on genetic algorithm optimization, characterized in that: The specific steps include: S1. Initialize the process parameter set, including pressure, temperature, cutting speed and material microstructure parameters, use the morphological optimization control model to analyze the process parameters, and generate a preliminary optimization plan; S2, using self-organizing mapping to map the results of morphological optimization from high-dimensional parameter space to low-dimensional morphological feature space, and adjusting the process parameters through adaptive mutation strategy; S3, integrating the morphological features and process parameters after self-organizing mapping into the chromosome encoding of the genetic algorithm, performing multiple generations of iterations, and optimizing the matching between microstructure and macro process parameters; S4. In the process of genetic algorithm evolution, combined with the evolutionary game model, the process parameter combination is dynamically adjusted through competition and cooperation of production strategies, and the evolutionary strategy is adjusted according to real-time production environment data through the fuzzy logic control mechanism; S5. In crossover and mutation operations, evolutionary game and fuzzy logic control are used to coordinately regulate key process parameters, so that the optimization results can maintain stability and consistency under complex conditions; S6. After each iteration, the morphological optimization and self-organizing map model are used to evaluate the fitness, and the genetic algorithm parameters are adjusted by combining evolutionary game and fuzzy logic control; S7. After reaching the convergence condition or the process target reaches the predetermined standard, the individual with the highest fitness is output as the final production process parameter and applied to actual production.

2. The method for producing micro spring washers based on genetic algorithm optimization according to claim 1, characterized in that: The S1 comprises the following steps: S11, set the initial process parameter set X = {P, T, v, M, σ, ∈, ξ}, where P is the pressure parameter, T is the temperature parameter, v is the cutting speed parameter, M is the material microstructure parameter, σ is the yield strength, ∈ is the strain characteristic, and ξ is the morphological adaptability factor, which is used to measure the influence of the process parameters on the microstructure morphology; S12, establish a morphology-process coupling optimization model, couple the microstructure morphology parameters with the process parameter set X, and calculate the morphology adaptability factor ξ, which is used to represent the comprehensive influence of different process parameters on the microstructure morphology; S13, according to the morphology-process coupling optimization model, the optimization target value ξ0 is set, and through iterative calculation, each parameter in the process parameter set X is adjusted to achieve the best adaptability with the target value ξ0, so as to optimize the microstructure morphology characteristics; S14, performing sensitivity analysis on the optimized process parameter set X, introducing a disturbance factor δξ according to the change of the morphological adaptability factor ξ, evaluating its influence on the microstructure morphology by adjusting the process parameter set, and performing parameter correction; S15, repeat the optimization and sensitivity analysis steps until the morphological adaptability factor ξ reaches the best match with the target value ξ0, and finally generate the optimized process parameter set X'', which is used as the preliminary optimization scheme and input into the subsequent optimization steps of the genetic algorithm.

3. The method for producing micro spring washers based on genetic algorithm optimization according to claim 1, characterized in that: The S2 comprises the following steps: S21, take the morphologically optimized process parameter set X' as input, construct a self-organizing mapping network, define the input layer of the network as the high-dimensional process parameter X', and the output layer as the low-dimensional morphological feature space Z = z1, z2, ..., z n }, where each eigenvalue z in Z i Represents the projection of morphological features in low-dimensional space; S22. Based on the nonlinear coupling relationship between the microstructure of the spring gasket material and the process parameters, a nonlinear mapping formula of the self-organizing map is constructed to map the process parameters X' to the low-dimensional morphological feature space Z: Among them, w ij is the weight of the mapping network, α j and β j is the nonlinear coefficient of the control process parameters, γ j is the morphological adaptability factor coefficient, δ j is the nonlinear oscillation adjustment factor, λ j is the frequency control factor, is the derivative of the morphological adaptability factor with respect to the process parameters; S23, by training the self-organizing map network, optimize the network weight W and control coefficient α j , β j , γ j , δ j and λ j , minimize the error function ε: in, is the expected morphological feature value, λ1 is the sensitivity matching factor, and λ2 is the weight regularization factor used to prevent overfitting; S24. Aiming at the nonlinear behavior in the mapping process, a multi-scale morphological feature adjustment strategy is used to grade the main and secondary morphological features; S25. Combining the morphological feature gain adjustment and the process parameter sensitivity, each parameter in the process parameter set X" is mutated and adjusted by adaptive mutation: Among them, κ i is to adjust the morphological feature gain coefficient, η is the mutation rate control factor; S26, iteratively adjusting and optimizing the process parameter set X'' until the optimization target is met, and finally taking the adjusted process parameter set X''' as the final optimization result and inputting it into the subsequent steps of the genetic algorithm.

4. The method for producing micro spring washers based on genetic algorithm optimization according to claim 1, characterized in that: S3 includes the following steps S31, the morphological feature set Z after self-organizing mapping processing is n |Integrated with the process parameter set X'' into the chromosome encoding C = {Z, X''} of the genetic algorithm; S32, initialize the population P = {C1, C2, ..., C k }, where each individual C i Represents different combinations of morphological characteristics and process parameters; the diversity of the population is achieved by perturbing the morphological characteristics and process parameters, so that the initial population has a wide distribution between the morphological characteristics and process parameters; S33. Define fitness function To evaluate the quality of each chromosome; the fitness function is formulated according to the comprehensive optimization goal of morphological characteristics and process parameters: Among them, ω1 and ω2 are weight factors, α i and β j is the nonlinear adjustment coefficient, and are the desired morphological characteristic values ​​and process parameter values, respectively; S34. In the crossover operation of the genetic algorithm, a hierarchical crossover strategy is adopted to perform a crossover operation on the morphological feature subset Z and a crossover operation on the process parameter subset X'': C new =α·C parent1 +(1-α)·C parent2 +γ·sin(π·(C parent1 -C parent2 )); Among them, C new is the newly generated offspring chromosome, α is the crossover rate control factor, γ is the nonlinear crossover adjustment factor, sin(π·(C parent1 -C parent2 )) is the introduced nonlinear interference term; S35. In the mutation operation, adaptive mutation strategies are respectively implemented for the morphological feature subset Z and the process parameter subset X'' in the chromosome, and the mutation rate η is dynamically adjusted according to the fitness distribution of the current population; S36. After multiple generations of iterations, monitor the fitness function The chromosome with the highest fitness is selected to enter the next generation until the fitness function converges or reaches the preset optimization goal. The final output chromosome C opt Including the optimized morphological features Z opt and process parameters X”’ as the optimal process configuration for the production of micro spring washers.

5. The method for producing micro spring washers based on genetic algorithm optimization according to claim 1, characterized in that: The S4 comprises the following steps: S41. In the evolution process of genetic algorithm, an evolutionary game model is constructed, each chromosome C=Z,X”'} is regarded as a game body, and its strategy space is set to S={s1,s2,...,s m }, where each strategy s i Corresponding to a process parameter adjustment plan; S42. Define the payoff function U(C,s i ), the payment function is based on the optimization objectives of the process parameters and the adaptability of the morphological characteristics: Among them, ω3 and ω4 are weight factors, λ is the adjustment coefficient, is the fitness function of the chromosome, and the payment function is used to evaluate the specific strategy s i The optimization effect under S43. In each generation of evolution, the game body selects strategies through the evolutionary game model to simulate the process of competition and cooperation; the strategy selection adopts a mixed strategy game model, in which each chromosome adjusts its process parameter combination according to the game results: Where η is the learning rate is the partial derivative of the payment function with respect to the strategy. After the strategy is updated, the process parameter combination of chromosome C is adjusted; S44, combined with the fuzzy logic control mechanism, the strategy space S of the evolutionary game is dynamically adjusted through the input of real-time production environment data; S45, in the process of evolution, chromosomes are adjusted through the payment function U(C,s i ) and fitness function The combined effect of the two factors is to continuously optimize the process parameter combination and select the strategy combination with the highest fitness to enter the next generation of iteration; S46, through multiple iterations of evolutionary game and fuzzy logic control mechanism, the optimized chromosome C is generated opt , C opt Contains the optimal morphological feature Z opt and process parameter combination X”” and serve as the optimal configuration solution for subsequent production processes.

6. The method for producing micro spring washers based on genetic algorithm optimization according to claim 1, characterized in that: The S5 comprises the following steps: S51. In the crossover operation of the genetic algorithm, for the morphological feature subset of chromosomes Z = z1, z2, ..., z n }Perform multi-point crossover and form a new chromosome Z by setting the crossover points new ; S52, in the crossover operation of the process parameter subset X'', the crossover rate is dynamically adjusted by fuzzy control, and the crossover operation is adjusted according to the real-time production environment; S53. After completing the crossover operation, the process parameters are further optimized using the evolutionary game model; each chromosome is regarded as a game subject, and the optimal combination of process parameters is achieved by adjusting the strategy space. The strategy update formula is: Where η is the learning rate factor, is the sensitivity of process parameters to strategy, cos(π·s j ) is the periodic adjustment factor; S54, when performing mutation operation, a multi-level dynamic mutation strategy is adopted to perform adaptive mutation on the morphological feature Z and process parameter X'' in the chromosome respectively: Among them, γ i is the variation enhancement coefficient of morphological characteristics, β i is the morphological feature variation suppression coefficient, λ j is the process parameter variation enhancement coefficient, α j is the process parameter variation suppression coefficient, and by adjusting these parameters, the direction and amplitude of variation can be controlled; S55. In each iteration, the mutation rate and mutation direction are adjusted according to the real-time production environment data through the fuzzy logic control mechanism; S56, the results of comprehensive evolutionary game and fuzzy logic control, combined with fitness function and the payment function U(C,s i ), evaluate the newly generated chromosomes, and retain the chromosomes with the highest fitness and stability to enter the next generation; S57, through multiple rounds of iterative optimization, the optimal chromosome C is finally generated opt , C opt Chromosomes contain morphological features that exhibit optimal stability and consistency under complex production conditions. opt and process parameter combination X””.

7. The method for producing micro spring washers based on genetic algorithm optimization according to claim 1, characterized in that: The S6 comprises the following steps: S61. After each iteration, the fitness of all chromosomes C = {Z, X'} is evaluated. The fitness evaluation is based on the joint feedback of the morphological optimization model and the self-organizing map type; the fitness function Combining the coupling relationship between morphological characteristics and process parameters, the nonlinear weighted formula is used to calculate: Among them, ω1 and ω2 are the weight coefficients of morphological characteristics and process parameters, α i and β j is the nonlinear adjustment factor, and are the target morphological characteristic values ​​and process parameter values; S62. Based on the fitness evaluation results, the key parameters of the genetic algorithm are dynamically adjusted; the adjustment formula of the crossover rate α is based on the interactive influence of morphological characteristics and process parameters, as follows: Among them, γ is the adjustment coefficient, θ is the phase adjustment factor, is the change in fitness function; S63, the mutation rate η is adjusted in real time, combined with the fuzzy logic control mechanism, and the frequency and intensity of the mutation operation are adjusted through the feedback of environmental data. The adjustment formula of the mutation rate is: Among them, μ is the fuzzy control gain coefficient, λ is the adjustment factor, fuzz(E) is the fuzzification processing result of environmental data, is the fitness change; S64. In each iteration, the competition and cooperation between different chromosomes are simulated through the evolutionary game model, and the strategy of dynamically adjusting the process parameters is: Among them, ζ is the strategy adjustment coefficient, κ i It is the strategy adjustment factor. Through the combination of sine and hyperbolic cosine functions, the strategy adjustment can better adapt to the complex nonlinear relationship between process parameters. S65. According to the feedback from the morphological optimization model and the self-organizing map model, chromosomes with low fitness are eliminated, and at the same time, some chromosomes are re-mutated in combination with the fuzzy logic control mechanism to maintain the diversity of the population and the global search capability; S66, after multiple generations of iterations, the chromosome C with the highest fitness and the best stability was finally selected opt , which contains the optimized morphological features Z opt and process parameter combination X”” to provide the best production process configuration solution; S67, the optimal chromosome C opt Applied to the actual production process, and continuously monitor and adjust the optimization parameters during the production process.

Images