Intelligent calculation method for jco reduction based on elastic-plastic theory and genetic algorithm

Abstract

This invention discloses an intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm, belonging to the field of sheet metal forming technology. Addressing the problems of existing technologies, a JCO forming process parameter model is constructed. A single-pass forming calculation model is established based on elastoplastic theory. By iteratively solving the bending angle convergence condition, the theoretical reduction, actual reduction after springback, and residual stress of a single pass are calculated. A multi-pass cumulative effect model is constructed by cascading multiple forming processes. Samples are generated within the compensation design space using the Latin hypercube sampling method. Combining experimental data with the simulation results of the multi-pass cumulative effect model, a response surface model of compensation and forming accuracy is constructed. With the optimization target parameter as a constraint, the response surface model is optimized using a genetic algorithm to obtain the optimal compensation amount. The optimal reduction is then calculated in conjunction with the theoretical reduction. This invention has high adaptability and eliminates the need for extensive sample testing, effectively saving time and materials.

Description

Technical Field

[0001] This invention belongs to the field of sheet metal plastic forming technology, specifically involving an intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm. Background Technology

[0002] JCO forming process is a key technology for manufacturing large-diameter straight seam steel pipes and ring parts. Its principle is to gradually form a flat plate into an open ring (J-shaped → C-shaped → O-shaped) through multiple progressive bending passes. For composite plates composed of a cladding layer (such as S304L stainless steel) and a base layer (such as Q235B carbon steel), due to the significant difference in the elastic modulus and yield strength of the two materials, uneven springback will occur during the forming process, making it difficult to control the precision indicators such as the ovality and opening degree of the final workpiece.

[0003] In existing technologies, the calculation of reduction amount mostly relies on empirical formulas or finite element simulations: empirical formulas ignore the elastic-plastic coupling of materials and the cumulative effect of multiple passes, resulting in low accuracy; although finite element simulations have higher accuracy, the calculation is time-consuming (a single pass simulation can take several hours), making it difficult to use for online process optimization. In addition, the work hardening of composite plates and the superposition of residual stress further increase the complexity of the reduction amount calculation, and traditional methods cannot meet the requirements of high-precision forming.

[0004] Therefore, there is an urgent need for a method to calculate the reduction amount that integrates theoretical modeling and intelligent algorithms, so as to improve the calculation efficiency while ensuring accuracy and achieve precise control of the JCO forming process. Summary of the Invention

[0005] To address the problems of low accuracy, poor efficiency, and difficulty in adapting to multi-pass forming of composite plates in existing technologies, this invention provides an intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm.

[0006] To achieve the above objectives, the present invention employs the following technical solutions:

[0007] The JCO reduction intelligent calculation method based on elastoplastic theory and genetic algorithm includes the following steps:

[0008] S1: Construct a JCO forming process parameter model, which includes material property parameters, process condition parameters, and optimization target parameters;

[0009] The material performance parameters include the elastic modulus of the coating. Base elastic modulus Coating yield strength Base layer yield strength Initial value of residual stress in the cladding layer Initial value of residual stress in the base layer The process parameters include the total thickness of the composite plate. Coating thickness punch radius Die radius Distance between the centers of the die Forming passes Central layer curvature The optimization target parameters include the target ellipticity. and target opening ;

[0010] S2: Based on the elastoplastic theory, a single-pass forming calculation model is established. By iteratively solving the bending angle convergence condition, the theoretical reduction, actual reduction after springback, and residual stress of a single pass are calculated.

[0011] The iterative process of the single-pass forming calculation model in step S2 includes:

[0012] S2.1: Initialize the curvature of the central layer Distance between the centers of the die This refers to the distance from the material layer where yielding begins to occur to the neutral layer. Since JCO forming involves large deformation, the thinner cladding layer undergoes plastic deformation, while the base layer undergoes elastoplastic deformation. The distance from the material layer to the neutral layer where yielding occurs in the base layer compression zone is also considered. : ;

[0013] The distance from the material layer where yielding occurs in the tensile zone to the neutral layer : ;

[0014] in This represents the neutral layer offset when the composite plate cross-section is in a purely elastic state. and The yield strengths of layers C and D are not equal due to the thickness of the plate, and can be measured experimentally.

[0015] S2.2: According to the fundamental assumption of pure bending in mechanics of materials, the axial internal forces in the bending section of the composite plate must satisfy static equilibrium, that is, the resultant force of the integral of the normal stress on the entire cross section is 0. The following stress equilibrium equation of the section can be obtained: ;

[0016] in, It is the distance (mm) from a point on the cross-section to the center layer; It is the curvature of the neutral layer;

[0017] The neutral layer offset of the bimetallic composite plate in its purely elastic state is obtained from the above formula. : ;

[0018] S2.3: Based on the equilibrium condition that the resultant force of the stress integral over the entire cross-section is 0, an equation can be constructed to solve for the neutral layer offset after the full plastic deformation of the composite panel cladding and the elastic-plastic deformation of the base layer. . ;

[0019] Calculate forming torque by region : ;

[0020] When a sheet metal enters an elastic-plastic bending state, and the external load is removed, the elastic portion disappears while the plastic portion remains. The sheet metal cannot fully recover its elastic-recovery strain. With elastic stress The expressions are as follows: ; ;

[0021] in, The distance (mm) from a point on the thickness section to the neutral layer. The radius of curvature of the neutral layer before unloading (mm). , The radius of curvature of the neutral layer after unloading; The elastic modulus of the material;

[0022] Elastic bending moment for: ;

[0023] in, This indicates the elastic stress of the coating material; Indicates the elastic stress of the base material;

[0024] Elastic bending moment equal to the bending moment at loading ,Right now The radius of curvature of the neutral layer after unloading was obtained. The formula is: ; ;

[0025] If the bending angle before rebound is As a neutral layer that is neither subjected to tension nor compression, the arc length before and after rebound is equal, that is... Calculate the bend angle after rebound, i.e., the target bend angle. ;

[0026] The step S2.3, which calculates the forming torque by region, includes the integral calculation of the compressive plastic torque of the coating, the compressive plastic torque of the base layer, the tensile elastic torque of the base layer, the tensile plastic torque of the base layer, and the compressive elastic torque of the base layer.

[0027] S2.4: Determine the bending angle after springback Is it smaller than the bend angle before rebound? If the conditions are not met, adjust the curvature of the central layer. Repeat steps S2.1-S2.3 until the springback bend angle is reached. Greater than the springback bending angle When convergence is reached, the theoretical reduction amount is output. and actual compression after rebound ; ; ;

[0028] S2.5: Calculate the theoretical reduction, actual reduction after springback, and residual stress based on the convergence results. , .

[0029] S3: Connect the multi-pass forming process and construct a multi-pass cumulative effect model. The cumulative effect model considers the superposition of work hardening and residual stress and outputs the characteristic parameters of the workpiece after multi-pass forming.

[0030] S3.1: Theoretical reduction, actual reduction after springback, and residual stress output from the single-pass forming calculation model. , As input, multiple forming processes are connected in sequence according to the number of passes. The number of JCO forming passes is generally an odd number of passes, ranging from 13 to 27. For products with special requirements, the number of passes may be even higher.

[0031] S3.2: Number of passes At that time, the 1st and Initial passes "J" and "C" are unaffected by work hardening of other passes, while all other passes are affected by work hardening of adjacent passes. A work hardening coefficient is introduced for passes other than the first and second passes. Coating yield strength of each pass and base layer yield strength Perform incremental corrections for the first and Coating yield strength of each pass Base yield strength Reset to initial value;

[0032] S3.3: Introduce a residual stress accumulation factor (the factor is adjustable) to superimpose and correct the residual stress of each pass;

[0033] S3.4: Calculation of workpiece based on pass-by-pass symmetric pairing rules The ellipticity is calculated based on a characteristic diameter and the initial design diameter, and the aperture is calculated based on the standard deviation of the reduction amount. (Ellipticity calculation follows.) The formula is: ;

[0034] in, For the maximum diameter, For the minimum diameter, This is the initial design diameter;

[0035] aperture The formula is: ;

[0036] in, This is the straight-line distance between the open ends of the tube blank.

[0037] If the number of passes is Then the first Each stage and the first Each pass is paired. If the number of passes is 13, then the 1st pass is paired with the 12th pass, the 2nd pass with the 11th pass, the 3rd pass with the 10th pass, the 4th pass with the 9th pass, the 5th pass with the 8th pass, and the 6th pass with the 7th pass. The characteristic diameter is the sum of the outer radii corresponding to each pair of passes. Based on this, the ellipticity and aperture are calculated.

[0038] S4: The Latin hypercube sampling method is used to generate samples in the compensation amount design space. Combined with experimental data and simulation results of the multi-channel cumulative effect model, a response surface model of compensation amount and forming accuracy is constructed.

[0039] The process of constructing the response surface model in step S4 includes:

[0040] S4.1: The compensation design space is divided into two groups based on whether the passes are affected by work hardening and residual stress. The principle of this grouping is as follows: The first group is the first pass of J and C, which is not affected by work hardening and residual stress in other passes, and a small amount of compensation is given to the reduction. The second group is affected by work hardening and residual stress in adjacent passes, and the springback is reduced, so negative compensation is given to the reduction. The values ​​of the compensation range for the two groups are based on values ​​obtained from multiple simulations and experiments.

[0041] S4.2: Import experimental data and merge experimental data with simulated sample points. Since experimental data can more realistically reflect the forming situation, double the weight is given to the experimental data.

[0042] S4.3: Use Latin hypercube sampling to generate 50 sample points, and call the multi-channel cumulative effect model to calculate the ellipticity and aperture of each sample point;

[0043] S4.4: A quadratic response surface model is used to approximate the nonlinear relationship between the compensation amount and the forming accuracy. The model form is as follows:

[0044] Ellipticity response surface model: ;

[0045] Aperture response surface model: ;

[0046] In the formula, , , , and , , , These are all regression coefficients, solved using the least squares method; This indicates the compensation amount for the nth passage; Indicates the first The compensation amount for each pass; the model fitting accuracy is determined by the coefficient of determination. Assessment, requirements To ensure model reliability, the closer the model is to 1, the better the model fit.

[0047] S5: Using the target parameters as constraints, the response surface model is optimized based on the genetic algorithm to obtain the optimal compensation amount, and the optimal reduction amount is calculated in combination with the theoretical reduction amount.

[0048] Step S5, which involves optimizing the response surface model using a genetic algorithm, includes:

[0049] S5.1: Employing an optimization objective function As a constraint, a smaller fitness value indicates a better optimization effect; the optimization objective function is... The formula is: ;

[0050] in: This represents the predicted ellipticity value from the response surface model. This represents the predicted aperture value from the response surface model. Target ellipticity; The target opening degree.

[0051] S5.2: Set the population size to 50 and the maximum number of iterations to 100, with the two sets of compensation ranges in step S4 as constraints;

[0052] S5.3: The compensation amount corresponding to the minimum value of the objective function is searched using a genetic algorithm, which is the optimal compensation amount. ;

[0053] S5.4: The optimal compensation amount is superimposed with the theoretical reduction amount to obtain the optimal reduction amount. : ;

[0054] in, This represents the theoretical reduction.

[0055] Compared with the prior art, the present invention has the following advantages:

[0056] High accuracy: Based on the elastoplastic theory, a single-pass model is established, taking into account the material nonlinearity and springback mechanism, and combined with the multi-pass cumulative effect (work hardening, residual stress superposition), which improves the accuracy by more than 30% compared with empirical formulas.

[0057] High efficiency: By replacing time-consuming multi-pass simulations with response surface models, the optimization process is shortened from several days to hours, meeting the needs of online process optimization;

[0058] Intelligentization: By integrating experimental and simulated data to construct a proxy model, and using genetic algorithms to achieve global optimization in the high-dimensional compensation space, the reliance on human experience is reduced.

[0059] Highly adaptable: The grouping compensation strategy is designed based on the characteristics of composite board materials, which can flexibly adjust material parameters and process conditions, and is suitable for JCO forming of composite boards of different specifications. Attached Figure Description

[0060] Figure 1 This is a schematic diagram of the cross-section of a bimetallic composite plate;

[0061] Figure 2 Schematic diagram of JCO forming principle for bimetallic composite plate;

[0062] Figure 3 The flowchart shows the intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm.

[0063] Figure 4 An open ring formed after 13 passes of composite board forming;

[0064] Figure 5 (a) Comparison before and after experimental optimization; (b) Scatter plot of opening distance and diameter data after optimization. Detailed Implementation

[0065] To gain a deeper understanding of this invention, we will provide a comprehensive and detailed description. However, this invention has various implementations and is not limited to the specific examples listed herein. These examples are presented to enhance a full understanding of the disclosure of this invention.

[0066] The JCO reduction intelligent calculation method based on elastoplastic theory and genetic algorithm includes the following steps:

[0067] S1: Construct a JCO forming process parameter model, which includes material property parameters, process condition parameters, and optimization target parameters;

[0068] The material performance parameters include the elastic modulus of the coating. Base elastic modulus Initial yield strength of the cladding layer Initial yield strength of the base layer Initial value of residual stress in the cladding layer Initial value of residual stress in the base layer The process parameters include the total thickness of the composite plate. Coating thickness punch radius Die radius Distance between the centers of the die Forming passes Initial curvature The optimization target parameters include the target ellipticity. and target opening A schematic diagram of the composite panel cross-section is shown below. Figure 1 As shown, the stress state of the composite panel varies with the curvature of the panel. This analysis combines the stress states of the three regions P1, P2, and P3 with the stress states of the four surfaces A, B, C, and D to analyze the complex stress situation of the thick cross-section under bending conditions, making the analysis clearer and more systematic.

[0069] S2: Based on the elastoplastic theory, a single-pass forming calculation model is established. By iteratively solving the bending angle convergence condition, the theoretical reduction, actual reduction after springback, and residual stress of a single pass are calculated.

[0070] Taking the first pass as an example, initialize the curvature of the central layer. The distance from the material layer to the neutral layer when the base layer undergoes elastic-plastic deformation and yields in the base layer compression zone. : The distance from the material layer where yielding occurs in the tensile zone to the neutral layer. : .calculate , , Solving by quadratic equations Integral calculation of torque Iterative adjustment Up to 0.0162, at this point (satisfy ), output , , , .

[0071] S3: Connect the multi-pass forming process and construct a multi-pass cumulative effect model. The cumulative effect model considers the superposition of work hardening and residual stress and outputs the characteristic parameters of the workpiece after multi-pass forming.

[0072] 13 courses in a row, courses 2-6 , (n is the path number); 7th path , Reset; residual stress accumulates at a rate of 1.1.

[0073] S4: The Latin hypercube sampling method is used to generate samples in the compensation amount design space. Combined with experimental data and simulation results of the multi-channel cumulative effect model, a response surface model of compensation amount and forming accuracy is constructed.

[0074] The process of constructing the response surface model in step S4 includes:

[0075] S4.1: The compensation amount design space is grouped by pass, as shown in the JCO forming principle diagram of the bimetallic composite plate. Figure 2 As shown; the first group (passes 1 and 7) is the first pass of J and C. The rebound amount has a small error compared with the theoretical value, so the compensation range is small and set to [-0.5, 0.5] mm. The second group (passes 2-6 and 8-12) is mainly negative compensation, and the compensation range is set to [-1, 0] mm.

[0076] S4.2: Import experimental data and merge experimental data with simulated sample points. Since experimental data can more realistically reflect the forming situation, double the weight is given to the experimental data.

[0077] S4.3: 50 sample points are generated using Latin hypercube sampling. The real samples generated by experiments and simulations in S4.2 are fused with the sample points generated by Latin hypercube sampling. The multi-channel cumulative effect model is called to calculate the ellipticity and aperture corresponding to each sample point. Sample points with large differences between ellipticity and aperture and the target value are removed, while sample points with small differences are saved to achieve the constraint of sample point data.

[0078] S4.4: A quadratic response surface model is used to approximate the nonlinear relationship between the compensation amount and the forming accuracy. The model form is as follows:

[0079] Ellipticity response surface model: ;

[0080] Aperture response surface model: ;

[0081] In the formula, , , , and , , , These are all regression coefficients, solved using the least squares method. This indicates the compensation amount for the nth passage; Indicates the first The compensation amount for each passage is approximated by a quadratic polynomial consisting of a constant term, a linear term, a quadratic term, and an interaction term, to represent the complex relationship between the compensation amount for all 13 passages and the ellipticity / aperture. The model fitting accuracy is determined using the coefficient of determination. Assessment, requirements To ensure model reliability, the closer the value is to 1, the better the model fit. Ellipticity response surface model. coefficient of determination =0.92, aperture response surface model coefficient of determination =0.89.

[0082] S5: Using the target parameters as constraints, the response surface model is optimized based on the genetic algorithm to obtain the optimal compensation amount, and the optimal reduction amount is calculated in combination with the theoretical reduction amount.

[0083] With a population size of 50 and 100 generations, the objective function converges to 0.03, indicating the optimal compensation amount. =[0.2,-0.3,-0.5,...,0.1] (13 dimensions), optimal reduction amount Verification =2.05%, =9.8°, meeting the target requirement. The entire process of the JCO intelligent calculation method based on elastoplastic theory and genetic algorithm is as follows: Figure 3 As shown;

[0084] By utilizing the optimal solution found through the algorithm, the formed open annulus is visualized, intuitively displaying the workpiece morphology and opening parameters, such as... Figure 4 As shown, the formed ring is symmetrical, with the opening centered and no obvious elliptical deformation. 1-13 shows the processing sequence and position of the JCO forming process. By breaking down the complex ring contour into multiple small arcs through 13 passes, the deformation risk of single-pass processing is reduced, and the contour accuracy is improved.

[0085] The optimized open-ring profile was designed using a MATLAB visualization program. By associating parameters such as pass order, angle, and optimized reduction amount with the geometric profile through MATLAB, the optimized forming effect can be quickly and intuitively displayed, significantly reducing the cost of physical trial and error.

[0086] S6: Comparative Analysis of JCO Forming Experiments. JCO forming experiments were conducted on the two reduction schemes before and after optimization, as follows: Figure 5 As shown, using the same reduction amount results in severe edge misalignment. Although edge misalignment still exists when using the optimized reduction amount, the forming quality is significantly improved compared to using a fixed reduction amount in 13 passes.

[0087] Contents not described in detail in this specification are prior art known to those skilled in the art. Although illustrative specific embodiments of the invention have been described above to facilitate understanding by those skilled in the art, it should be understood that the invention is not limited to the scope of the specific embodiments. Various modifications are readily apparent to those skilled in the art as long as they fall within the spirit and scope of the invention as defined and determined by the appended claims, and all inventions utilizing the concept of this invention are protected.

Claims

1. A JCO reduction intelligent calculation method based on elastoplastic theory and genetic algorithm, characterized in that, The method includes the following steps: S1: Construct a JCO forming process parameter model, which includes material property parameters, process condition parameters, and optimization target parameters; S2: Based on the elastoplastic theory, a single-pass forming calculation model is established. By iteratively solving the bending angle convergence condition, the theoretical reduction, actual reduction after springback, and residual stress of a single pass are calculated. S3: Connect the multi-pass forming process and construct a multi-pass cumulative effect model. The cumulative effect model considers the superposition of work hardening and residual stress and outputs the characteristic parameters of the workpiece after multi-pass forming. S4: The Latin hypercube sampling method is used to generate samples in the compensation amount design space. Combined with experimental data and simulation results of the multi-channel cumulative effect model, a response surface model of compensation amount and forming accuracy is constructed. S5: Using the target parameters as constraints, the response surface model is optimized based on the genetic algorithm to obtain the optimal compensation amount, and the optimal reduction amount is calculated in combination with the theoretical reduction amount.

2. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 1, characterized in that, In step S1, the material performance parameters include the coating elastic modulus. Base elastic modulus Coating yield strength Base layer yield strength Initial value of residual stress in the cladding layer Initial value of residual stress in the base layer The process parameters include the total thickness of the composite plate. Coating thickness punch radius Die radius Distance between the centers of the die Forming passes Central layer curvature The optimization target parameters include the target ellipticity. and target opening .

3. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 2, characterized in that, The iterative process of the single-pass forming calculation model in step S2 includes: S2.1: Initialize the curvature of the central layer Distance between the centers of the die The distance from the material layer where yielding begins to occur to the neutral layer; the distance from the material layer where yielding occurs in the base compression zone to the neutral layer. : ; The distance from the material layer where yielding occurs in the tensile zone to the neutral layer : ; in, This represents the neutral layer offset when the composite plate cross-section is in a purely elastic state. and The yield strengths of layers C and D were measured experimentally. S2.2: Solving for neutral layer offset using the cross-sectional stress balance equation The formula for the stress equilibrium equation of the cross section is: ; in, It is the distance from a point on the cross-section to the central layer; It is the curvature of the neutral layer; The neutral layer offset of the bimetallic composite plate in its purely elastic state is obtained from the above formula. The formula is: ; S2.3: Based on the equilibrium condition that the integral resultant force of the normal stress over the entire cross-section is 0, an equation is constructed to solve for the neutral layer offset after the full plastic deformation of the composite cladding and the elastic-plastic deformation of the base layer. : ; Calculate forming torque by region : ; After the sheet metal enters the elastoplastic bending state, its elastic-recovery strain With elastic stress The expressions are as follows: ; ; in, The distance from a point on the thickness section to the neutral layer. The radius of curvature of the neutral layer before unloading. , The radius of curvature of the neutral layer after unloading; The elastic modulus of the material; Elastic bending moment for: ; in, This indicates the elastic stress of the coating material; Indicates the elastic stress of the base material; Elastic bending moment equal to the bending moment at loading ,Right now The radius of curvature of the neutral layer after unloading was obtained. The formula is: ; ; If the bending angle before rebound is As a neutral layer that is neither subjected to tension nor compression, the arc length before and after rebound is equal, that is... Calculate the bend angle after rebound, i.e., the target bend angle. ; S2.4: Determine the bending angle after springback Is it smaller than the bend angle before rebound? If the conditions are not met, adjust the curvature of the central layer. Repeat steps S2.1-S2.3 until the springback bend angle is reached. Greater than the springback bending angle When convergence is reached, the theoretical reduction amount is output. and actual compression after rebound ; ; ; S2.5: Calculate the theoretical reduction, actual reduction after springback, and residual stress based on the convergence results. , .

4. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 3, characterized in that, The construction process of the multi-channel cumulative effect model in step S3 includes: S3.1: Theoretical reduction, actual reduction after springback, and residual stress output from the single-pass forming calculation model. , As input, multiple forming processes are connected in sequence according to the number of passes; S3.2: Number of passes At that time, the 1st and Initial passes "J" and "C" are unaffected by work hardening of other passes, while all other passes are affected by work hardening of adjacent passes. A work hardening coefficient is introduced for passes other than the first and second passes. Coating yield strength of each pass and base layer yield strength Perform incremental corrections for the first and Coating yield strength of each pass Base yield strength Reset to initial value; S3.3: Introduce a residual stress accumulation factor to superimpose and correct the residual stress of each pass; S3.4: Calculation of workpiece based on pass-by-pass symmetric pairing rules The ellipticity is calculated based on a characteristic diameter and the initial design diameter, and the aperture is calculated based on the standard deviation of the reduction amount. (Ellipticity calculation follows.) The formula is: ; in, For the maximum diameter, For the minimum diameter, This is the initial design diameter; aperture The formula is: ; in, This is the straight-line distance between the open ends of the tube blank.

5. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 4, characterized in that, The process of constructing the response surface model in step S4 includes: S4.1: The compensation design space is divided into two groups based on whether the passes are affected by work hardening and residual stress. The principle of this grouping is as follows: The first group is the first pass of J and C, which is not affected by work hardening and residual stress of other passes, and a small amount of compensation is given to the reduction; the second group is affected by work hardening and residual stress of adjacent passes, and the springback is reduced, so negative compensation is given to the reduction; the values ​​of the compensation range of the two groups are based on values ​​obtained from multiple simulations and experiments. S4.2: Import experimental data and merge the experimental data with the simulated sample points; S4.3: Use Latin hypercube sampling to generate 50 or more supplementary sample points. Merge the real samples generated by experiments and simulations in S4.2 with the sample points generated by Latin hypercube sampling. Call the multi-channel cumulative effect model to calculate the ellipticity and aperture corresponding to each sample point. Remove sample points with large differences between the ellipticity and aperture and the target value, and save sample points with small differences to achieve the constraint of sample point data. S4.4: A quadratic response surface model is used to approximate the nonlinear relationship between the compensation amount and the forming accuracy. The model form is as follows: Ellipticity response surface model: ; Aperture response surface model: ; In the formula, , , , and , , , These are all regression coefficients, solved using the least squares method; This indicates the compensation amount for the nth passage; Indicates the first The compensation amount for each pass; the model fitting accuracy is determined by the coefficient of determination. Evaluate.

6. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 5, characterized in that, Step S5, which involves optimizing the response surface model using a genetic algorithm, includes: S5.1: Employing an optimization objective function To constrain and optimize the objective function The formula is: ; in, This represents the predicted ellipticity value from the response surface model. This represents the predicted aperture value from the response surface model. Target ellipticity; For the target opening degree; S5.2: Set the population size and maximum number of iterations, using the two sets of compensation ranges in step S4 as constraints; S5.3: The compensation amount corresponding to the minimum value of the objective function is searched using a genetic algorithm, which is the optimal compensation amount. ; S5.4: The optimal compensation amount is superimposed with the theoretical reduction amount to obtain the optimal reduction amount. The formula is: ; in, This represents the theoretical reduction.

7. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 6, characterized in that, The step S2.3, which calculates the forming torque by region, includes the integral calculation of the compressive plastic torque of the coating, the compressive plastic torque of the base layer, the tensile elastic torque of the base layer, the tensile plastic torque of the base layer, and the compressive elastic torque of the base layer.

8. The intelligent calculation method for JCO reduction based on elastoplastic theory and genetic algorithm according to claim 7, characterized in that, The symmetrical pairing rule for the number of passes in step S3.4 is as follows: if the number of passes is... Then the first Each stage and the first Each lane is paired.

Images