A structural part non-uniform allowance machining deformation control process optimization method and application

Abstract

The application discloses a structural part non-uniform allowance machining deformation control process optimization method and application thereof, and the method comprises the following steps: obtaining an initial geometric model of a structural part, material properties and clamping boundary conditions, establishing a to-be-optimized allowance design domain and discretizing the to-be-optimized allowance design domain into material layers along a machining depth, and defining the reserved / removal state of each layer as a design variable; a machining deformation evaluation model considering the structural stiffness evolution in the material removal process is established, a target function and a constraint condition are constructed with the minimum machining deformation as the target, the sensitivity of the design variable is calculated by using the adjoint variable method and is subjected to spatial filtering, the design variable is updated relying on a bidirectional progressive optimization strategy, and the final non-uniform allowance scheme is obtained after iterative convergence. The allowance distribution problem is converted into a material distribution optimization problem, the machining deformation sensitive area can be effectively identified, the allowance is differentiated and distributed, the deformation control effect and the engineering manufacturability are considered, and the method is suitable for the finishing process of aerospace, die forgings and thin-walled precision structural parts.

Description

Technical Field

[0001] This invention relates to the field of precision manufacturing and machining deformation control technology, and in particular to an optimization method and application for controlling deformation during machining of non-uniform allowances in structural components. Background Technology

[0002] Aerospace structural components typically serve as critical load-bearing structures in aircraft, launch vehicles, and other equipment. Their machining quality directly impacts the overall structural strength, service reliability, and assembly accuracy. These components are highly susceptible to deformation after machining, affecting subsequent assembly and service performance. Therefore, effectively controlling residual stress release and deformation during component machining has always been a key technical challenge in high-precision aerospace manufacturing.

[0003] For large and complex structural components, different regions typically exhibit significantly different geometric characteristics and structural stiffness, while the internal residual stress field also displays a non-uniform spatial distribution. The allocation of material allowance directly affects the overall stiffness evolution and residual stress release path of the workpiece during machining, thus significantly influencing the final machining deformation. Machining allowance is not merely a machining parameter, but a crucial technological factor that simultaneously affects the stiffness distribution and residual stress field rebalancing process of the structural component; its allocation scheme directly relates to the effectiveness of machining deformation control.

[0004] Traditional processes typically employ a relatively conservative method of uniform allowance distribution. While this approach facilitates process design and implementation, it fails to adequately consider the differences in stiffness characteristics and residual stress distribution across different regions. This can easily lead to a mismatch between local stiffness and residual stress field distribution during processing, resulting in complex and unpredictable machining deformations in structural components during material removal. Consequently, it is difficult to meet the high-precision, low-deformation manufacturing requirements of aerospace structural components.

[0005] Compared to uniform allowance distribution, non-uniform allowance design can adjust local allowances in a targeted manner based on the stiffness characteristics, residual stress state, and deformation sensitivity of different regions of a structural component. This allows for a certain degree of coordination between the structural stiffness distribution and residual stress release behavior during processing, enabling proactive control of processing deformation. However, non-uniform allowance design typically involves combinations of allowances from multiple processing areas, multiple levels, or multiple local features. This results in a large number of design variables, a vast design space, and the coupling of allowance adjustments in different areas, leading to complex process decision-making and significant optimization challenges.

[0006] Therefore, there is an urgent need for a non-uniform allowance machining deformation control process that can combine the geometric state of structural components with the distribution of residual stress and realize the dynamic optimization allocation of machining allowance based on sensitivity analysis, so as to solve the technical problem of the difficulty in accurately and actively controlling the machining deformation of large and complex structural components, and meet the manufacturing needs of low deformation and high precision structural components in high-precision manufacturing fields such as aerospace. Summary of the Invention

[0007] The purpose of this invention is to provide a method and application for optimizing the machining deformation control of non-uniform machining allowances in structural components. Based on the initial geometry and residual stress field of the structural component to be optimized, the machining allowance region to be optimized is discretized according to groove characteristics and layers, and the retention or removal status of each local material layer is defined as a design variable. Under the current geometric and stress states, the displacement and stress fields of the structural component are obtained through a machining deformation prediction model, and the sensitivity of each design variable to the target machining deformation is calculated using the adjoint variable method. Based on the sensitivity results, design variables with less impact on deformation are dynamically reduced, and the design variables are iteratively updated bidirectionally by adding or removing them in conjunction with machining deformation until a preset convergence condition is met, thereby obtaining the final non-uniform machining allowance scheme and achieving active regulation and precise control of the machining deformation of the structural component.

[0008] To achieve the above objectives, the present invention provides a method for optimizing the process of controlling deformation during machining of non-uniform allowances in structural components, comprising the following steps: S1. Obtain the initial geometric model, material property information and clamping boundary conditions of the structural component before optimization. Based on the groove features, web area, rib area or other local features of the area to be processed of the structural component, establish the design domain of the allowance to be optimized.

[0009] The design domain for the margin to be optimized is discretized into several material layers along the machining depth direction, and each layer of material for each groove feature is defined as a design variable. For features including... Each groove feature is divided along the depth direction into several groove features. For structural components with layer allowances, the design variables can be expressed as: ; in, Indicates the first The first groove feature Layer material retained, Indicates the first The first groove feature Layer material removal.

[0010] Therefore, the problem of non-uniform margin optimization is transformed into a material distribution optimization problem.

[0011] S2. Based on the initial geometric model, material property information and clamping boundary conditions, establish a processing deformation evaluation model. With the minimum processing deformation of the structural part as the optimization objective, construct a processing deformation objective function and construct constraints that meet the upper and lower limits of processing allowance and the requirements of manufacturability. S3. Based on the processing deformation evaluation model, the sensitivity of each design variable to the processing deformation target is calculated using the adjoint variable method; S4. Perform spatial filtering on the sensitivity calculated in step S3 to obtain the filtered sensitivity. S5. Based on the filtered sensitivity results, a two-way progressive optimization strategy is adopted to delete, retain, or add and update the design variables to obtain a new allowance distribution state. The machining deformation of the structural parts under the new allowance distribution state is calculated. Among them, the machining allowance is reduced first in areas with low sensitivity and strong structural rigidity, and the machining allowance is retained first in areas with high sensitivity and easy structural deformation. S6. Repeat steps S3 to S5 until the preset convergence condition is met to obtain the final non-uniform allowance machining scheme for the structural component. That is, based on the updated design variables, re-establish the structural geometry and its deformation under the current allowance distribution state, and repeat the process of "deformation evaluation - sensitivity calculation - sensitivity filtering - design variable update" until the preset convergence condition is met.

[0012] Convergence criteria include at least one of the following: ① The state of the design variables remains unchanged during the continuous preset number of iterations; ②The relative change in the processing deformation objective function obtained from two adjacent iterations is less than the preset tolerance.

[0013] The relative change in the objective function between two consecutive iterations can be expressed as: ; in, This is the preset convergence tolerance. The convergence tolerance value is less than... .

[0014] Once the convergence condition is met, the final design variable distribution state is output, and the final non-uniform machining allowance scheme for the structural component is obtained accordingly.

[0015] Preferably, the sensitivity in step S3 is calculated using the adjoint variable method, specifically including: Under the current surplus distribution state, establish the static equilibrium equations for the structural components: ; in, The overall stiffness matrix of the structural components. Let be the nodal displacement vector. For load vector, This serves as the residual distribution state identifier for the current iteration step. The objective function of machining deformation is expressed as a function of design variables, and an adjoint variable with the same dimension as the displacement vector is introduced. Construct the Lagrangian function containing static equilibrium constraints: ; Taking the total differential of the Lagrange function, setting its partial derivatives with respect to the displacement variable to zero and satisfying the stationary condition, we obtain the adjoint equation: ; Solving the adjoint equation yields the adjoint variables; Based on the obtained accompanying variables, the sensitivity corresponding to each design variable is calculated. The sensitivity is used to characterize the degree of influence of the corresponding material layer change on the processing deformation target. Among them, the The sensitivity corresponding to the first design variable, while also including the first design variable. The changes in the local stiffness matrix and the equivalent nodal force changes caused by the state changes of the design variables. The sensitivity corresponding to each design variable can be expressed as: ; in, This represents the rate of change of the local stiffness matrix caused by changes in design variables. This represents the rate of change of the equivalent nodal load of residual stress caused by changes in design variables. Therefore, sensitivity simultaneously reflects the combined effect of local material variations on structural stiffness and residual stress release.

[0016] Preferably, step S4 performs spatial filtering on the sensitivity of each design variable calculated in step S3. The spatial filtering process obtains the filtered sensitivity by comprehensively considering the sensitivity of adjacent design variables within a preset filtering radius and the distance weight between adjacent design variables and the current design variable.

[0017] Sensitivity after filtering It can be represented as: ; in, For the position located at the A set of adjacent design variables within a filter radius of a design variable. These are distance-weighted coefficients used to characterize design variables. With design variables Neighborhood influence between them.

[0018] The distance weighting coefficient can be expressed as: ; in, For the filter radius, Design variables With design variables The center distance.

[0019] Preferably, the iterative update of design variables in step S5 adopts a two-way incremental optimization strategy, including: Update the design variables that are currently in the retained state and whose filtered sensitivity is lower than the deletion threshold to the removal state; Update the design variables that are currently removed and whose filtered sensitivity is higher than the added threshold to the retained state.

[0020] The update rule can be represented as: The design variables except for the state are restored to their reserved state. The update rule can be expressed as: ; in, For the first During the nth iteration The state of each design variable The deletion threshold, To add a threshold.

[0021] The threshold is dynamically determined based on the statistical characteristics of the sensitivity distribution in the current iteration. If the filtered sensitivity is approximately fitted to a statistical distribution, the deletion and addition thresholds are dynamically adjusted according to the sensitivity distribution characteristics and the changing trend of the objective function in the current iteration, so as to improve search efficiency in the early stage of optimization and enhance structural stability in the later stage of optimization.

[0022] During the design variable update process, the following constraints must be satisfied: processing sequence constraint, minimum processing allowance constraint, maximum processing allowance constraint, and manufacturability constraint. Specifically: The local allowance shall not be less than the minimum machining allowance; Local allowance shall not exceed the maximum permissible allowance; Material removal from adjacent layers satisfies hierarchical continuity constraints; Key geometric feature areas and easily deformable areas should be given priority in retaining necessary allowances; It meets the requirements of tool accessibility, machining path continuity, and actual clamping process.

[0023] Preferably, the final non-uniform machining allowance scheme for the structural component obtained in step S6 satisfies the following allowance allocation rules: For areas with low sensitivity, high structural rigidity, or weak influence from processing deformation, priority should be given to reducing processing allowance; For areas with high sensitivity, key geometric features, weak structures, or easily deformable structures, larger machining allowances should be reserved first. Preferably, the convergence condition in step S6 includes at least one of the following: the state of the design variables remains unchanged during a preset number of consecutive iterations, or the relative change of the processing deformation objective function obtained from two adjacent iterations is less than a preset tolerance.

[0024] Preferably, in step S1, when the design domain to be optimized is discretized into several material layers along the processing depth direction, the design variables corresponding to one or more adjacent material layers in a local area of ​​the structural component are merged into a single combined design variable; the single combined design variable is used to characterize the synchronous retention state or synchronous removal state of all material layers within the corresponding range.

[0025] Preferably, in step S1, the material property information includes basic mechanical property parameters and plasticity parameters; When the material property information also includes the residual stress state, the method for obtaining the residual stress field includes at least one of the following: The overall residual stress field of the structural component is directly measured using stress measurement methods. During the pre-processing of the blank, deformation force data reflecting the overall deformation trend of the blank are monitored. A residual stress field inference method based on deformation force is used to infer the residual stress field of the initial blank of the structural component and obtain the initial residual stress field distribution of the structural component.

[0026] Preferably, in step S2, the processing deformation evaluation model adopts the finite element prediction method or the neural network prediction method. The processing deformation evaluation model is used to simulate the residual stress release and structural deformation during the material removal process under different allowance distribution states.

[0027] The machining deformation prediction model is used to characterize the stiffness change, residual stress release and machining deformation of structural components during the material removal process under different combinations of design variables.

[0028] In each iteration state, the structural components satisfy the static equilibrium equations: ; in, To be related to design variables The relevant overall stiffness matrix, Let be the nodal displacement vector. This is the equivalent nodal load vector caused by the release of residual stress.

[0029] The overall stiffness matrix and the equivalent nodal load vector both change with the state of the design variables, which are used to characterize the stress and deformation characteristics of the structural members under different margin distribution conditions.

[0030] An objective function is constructed by taking one or more of the following as optimization objectives: minimizing the overall deformation of the structural component after processing, minimizing the displacement of key measuring points, and minimizing the error of key contours.

[0031] The objective function is to process deformation at key locations. It can be represented as: ; Or it can be expressed as the displacement deviation of key nodes: ; in, For the first Calculated displacement at key locations, For the target displacement value, This refers to the number of key positions.

[0032] When the objective is to minimize the final processing deformation, the objective function can also be written as: ; It also satisfies design variable value constraints, total margin constraints, and manufacturability constraints.

[0033] The clamping boundary conditions obtained in step S1 are set according to the actual machining and clamping state, and fixed displacement constraints are applied to the fixed position of the structural component.

[0034] This invention also provides an application of the optimization method for controlling deformation during machining of non-uniform allowance in structural components. The above-mentioned optimization method for controlling deformation during machining of non-uniform allowance in structural components is applied to the control of deformation during the finishing of aerospace structural components, die-forged structural components, or thin-walled precision structural components.

[0035] Therefore, the present invention employs the above-mentioned method and application for optimizing the machining deformation control process of non-uniform allowances in structural components, which has the following beneficial effects: (1) This invention transforms the problem of allocating machining allowance for structural components into a material distribution optimization problem that considers residual stress release and structural stiffness evolution, and can reveal the influence law of machining allowance on machining deformation from the mechanism level; (2) The present invention uses the adjoint variable method to calculate the sensitivity of design variables, which can efficiently identify key allowance areas that have a significant impact on processing deformation and reduce the optimization difficulty brought about by high-dimensional design space; (3) The present invention realizes the dynamic reduction and replenishment of design variables through a two-way incremental optimization strategy, which is more conducive to the active control of structural component processing deformation compared with the traditional uniform allowance process. (4) The present invention can take into account both the processing deformation control effect and engineering feasibility, and is applicable to the manufacturing process of large and complex thin-walled structural parts, die-forged structural parts and other high-precision structural parts in aerospace.

[0036] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0037] Figure 1 This is a flowchart of a process optimization method for controlling deformation during non-uniform allowance machining of structural components, as described in an embodiment of the present invention. Figure 2This is a schematic diagram of the deformation measurement location of the structural component in an embodiment of the present invention; Figure 3 This is a diagram showing the distribution of finishing allowances for the experimentally machined structural components in this embodiment of the invention. Figure 4 The following are comparison diagrams of deformation cloud maps of non-uniform allowance parts and uniform allowance parts in the experiment of the present invention: (a) Deformation cloud map of structural part 1; (b) Deformation cloud map of structural part 2. Detailed Implementation

[0038] Example like Figure 1 As shown, this invention discloses a method for optimizing the machining deformation control process of non-uniform allowance structural components, the steps of which include: S1. Obtain the initial geometric model, material properties, and clamping boundary conditions of the structural component before optimization. The material properties include basic mechanical and plastic parameters. Based on the groove features, web region, rib region, or other local features of the area to be processed, establish the design domain for the allowance to be optimized.

[0039] The design domain is discretized into several material layers along the processing depth direction, and each layer of material for each groove feature is defined as a design variable. For features containing Each groove feature is divided along the depth direction into several groove features. For structural components with layer allowances, the design variables can be expressed as: ; in, Indicates the first The first groove feature Layer material retained, Indicates the first The first groove feature Layer material removal.

[0040] Therefore, the problem of non-uniform margin optimization is transformed into a material distribution optimization problem.

[0041] When the design domain of the margin to be optimized is discretized into several material layers along the processing depth direction, the design variables corresponding to one or more adjacent material layers in the local area of ​​the structural component are merged into a single combined design variable; the single combined design variable is used to characterize the synchronous retention state or synchronous removal state of all material layers in the corresponding range.

[0042] When the material property information also includes the residual stress state, the method for obtaining the residual stress field of the structural component includes at least one of the following: The overall residual stress field of the structural component is directly measured using stress measurement methods. During the pre-processing of the blank, deformation force data reflecting the overall deformation trend of the blank are monitored. A residual stress field inference method based on deformation force is used to infer the residual stress field of the initial blank of the structural component and obtain the initial residual stress field distribution of the structural component.

[0043] The clamping boundary conditions are set according to the actual machining and clamping state, and fixed displacement constraints are applied to the fixed position of the structural component.

[0044] Specific parameters for this embodiment: The verification object is a 7075 aluminum alloy multi-cavity thin-walled die-forged structural part with external dimensions of 600mm×240mm×30mm and web thickness of 4mm; roughing is carried out by layer-by-layer milling with a cutting depth of 2mm per layer for a total of 9 layers, using a layer-priority machining strategy, and collecting 54 deformation force data to infer the residual stress field; the total allowance for semi-finishing and finishing is 2.0mm, with each 0.3mm divided into a design variable and the last 0.5mm combined into one, for a total of 36 design variables.

[0045] S2. Based on the initial geometric model, material properties, residual stress field and clamping boundary conditions, establish a machining deformation prediction model. With the minimum machining deformation of the structural part as the optimization objective, construct the machining deformation objective function and construct constraints that satisfy the upper and lower limits of machining allowance and the requirements of manufacturability.

[0046] The machining deformation prediction model adopts the finite element method or the neural network method. The machining deformation prediction model is used to simulate the residual stress release and structural deformation during the material removal process under different allowance distribution states.

[0047] In each iteration state, the structural components satisfy the static equilibrium equations: ; in, To be related to design variables The relevant overall stiffness matrix, Let be the nodal displacement vector. This is the equivalent nodal load vector caused by the release of residual stress.

[0048] The overall stiffness matrix and the equivalent nodal load vector both change with the state of the design variables, which are used to characterize the stress and deformation characteristics of the structural members under different margin distribution conditions.

[0049] An objective function is constructed by taking one or more of the following as optimization objectives: minimizing the overall deformation of the structural component after processing, minimizing the displacement of key measuring points, and minimizing the error of key contours.

[0050] The objective function is to process deformation at key locations. It can be represented as: ; Or it can be expressed as the displacement deviation of key nodes: ; in, For the first Calculated displacement at key locations, For the target displacement value, This refers to the number of key positions.

[0051] When the objective is to minimize the final processing deformation, the objective function can also be written as: ; It also satisfies design variable value constraints, total margin constraints, and manufacturability constraints.

[0052] S3. Based on the machining deformation prediction model, obtain the displacement and stress fields corresponding to the current allowance distribution state, and calculate the sensitivity of each design variable to the machining deformation target using the adjoint variable method; specifically including: Under the current surplus distribution state, establish the static equilibrium equations for the structural components: ; in, The overall stiffness matrix of the structural components. Let be the nodal displacement vector. For load vector, This serves as the residual distribution state identifier for the current iteration step. The objective function of machining deformation is expressed as a function of design variables, and an adjoint variable with the same dimension as the displacement vector is introduced. Construct the Lagrangian function containing static equilibrium constraints: ; Taking the total differential of the Lagrange function, setting its partial derivatives with respect to the displacement variable to zero and satisfying the stationary condition, we obtain the adjoint equation: ; Solving the adjoint equation yields the adjoint variables; Based on the obtained accompanying variables, the sensitivity corresponding to each design variable is calculated. The sensitivity is used to characterize the degree of influence of the corresponding material layer change on the processing deformation target. Among them, the The sensitivity corresponding to the first design variable, while also including the first design variable. The changes in the local stiffness matrix and the equivalent nodal force changes caused by the state changes of the design variables. The sensitivity corresponding to each design variable can be expressed as: ; in, This represents the rate of change of the local stiffness matrix caused by changes in design variables. This represents the rate of change of the equivalent nodal load of residual stress caused by changes in design variables. Therefore, sensitivity simultaneously reflects the combined effect of local material variations on structural stiffness and residual stress release.

[0053] S4. To suppress numerical instability phenomena such as finite element discretization errors, local stress singularities, and checkerboard patterns, the sensitivity calculated in step S3 is spatially filtered to obtain the filtered sensitivity. The spatial filtering process combines the sensitivities of adjacent design variables within a preset filtering radius with the distance weights between adjacent design variables and the current design variable to obtain the filtered sensitivity.

[0054] Sensitivity after filtering It can be represented as: ; in, For the position located at the A set of adjacent design variables within a filter radius of a design variable. These are distance-weighted coefficients used to characterize design variables. With design variables Neighborhood influence between them.

[0055] The distance weighting coefficient can be expressed as: ; in, For the filter radius, Design variables With design variables The center distance.

[0056] Spatial filtering is used to obtain a smoother sensitivity distribution that better reflects the actual processing characteristics, providing a basis for subsequent design variable updates.

[0057] S5. Based on the filtered sensitivity results, combined with the manufacturability constraints, processing sequence constraints and upper and lower limits of the allowance, a two-way progressive optimization strategy is adopted to delete, retain or add and update the design variables to obtain a new allowance distribution state, and the processing deformation of the structural parts under the new allowance distribution state is calculated. The design variable iterative update adopts a two-way incremental optimization strategy, including: Update the design variables that are currently in the retained state and whose filtered sensitivity is lower than the deletion threshold to the removal state; Update the design variables that are currently removed and whose filtered sensitivity is higher than the added threshold to the retained state.

[0058] ; in, For the first During the nth iteration The state of each design variable The deletion threshold, To add a threshold.

[0059] The threshold is dynamically determined based on the statistical characteristics of the sensitivity distribution in the current iteration. If the filtered sensitivity is approximately fitted to a statistical distribution, the deletion and addition thresholds are dynamically adjusted according to the sensitivity distribution characteristics and the changing trend of the objective function in the current iteration, so as to improve search efficiency in the early stage of optimization and enhance structural stability in the later stage of optimization.

[0060] During the design variable update process, the following constraints must be satisfied: processing sequence constraint, minimum processing allowance constraint, maximum processing allowance constraint, and manufacturability constraint. Specifically: The local allowance shall not be less than the minimum machining allowance; Local allowance shall not exceed the maximum permissible allowance; Material removal from adjacent layers satisfies hierarchical continuity constraints; Key geometric feature areas and easily deformable areas should be given priority in retaining necessary allowances; It meets the requirements of tool accessibility, machining path continuity, and actual clamping process.

[0061] S6. Repeat steps S3 to S5 until the preset convergence condition is met to obtain the final non-uniform allowance machining scheme for the structural component. That is, based on the updated design variables, re-establish the structural geometry, stress state, and finite element model under the current allowance distribution state, and repeat the process of "finite element analysis - sensitivity calculation - sensitivity filtering - design variable update" until the preset convergence condition is met.

[0062] Convergence criteria include at least one of the following: ① The state of the design variables remains unchanged during the continuous preset number of iterations; ②The relative change in the processing deformation objective function obtained from two adjacent iterations is less than the preset tolerance.

[0063] The relative change in the objective function between two consecutive iterations can be expressed as: ; in, This is the preset convergence tolerance. The convergence tolerance value is less than... .

[0064] Once the convergence condition is met, the final design variable distribution state is output, and the final non-uniform machining allowance scheme for the structural component is obtained accordingly (e.g., Figure 2 (As shown).

[0065] The final non-uniform machining allowance scheme for structural components shall satisfy the following allowance allocation rules: For areas with low sensitivity, high structural rigidity, or weak influence from processing deformation, priority should be given to reducing processing allowance; For areas with high sensitivity, key geometric features, weak structures, or easily deformable structures, a larger machining allowance should be reserved.

[0066] After finishing, release the floating clamping constraints at the four corners of the structural component, and use a machine tool probe to measure the deformation at preset measuring points (measurement positions are as follows). Figure 3 (As shown). In this embodiment, a non-uniform allowance part is compared with a traditional 2.0mm uniform allowance part. The measurement results show that at measuring point (580, 15), the deformation of the non-uniform allowance part is -0.106mm, while that of the uniform allowance part is -1.653mm; at measuring point (580, 225), the deformation of the non-uniform allowance part is 0.529mm, while that of the uniform allowance part is -1.176mm. The deformation amplitude of the non-uniform allowance part is significantly reduced.

[0067] Deformation cloud map drawn based on measurement results (e.g.) Figure 4 (As shown) further demonstrates that, Figure 4 (a) The overall deformation distribution of the structural component using the method of the present invention is more gradual, and the local large deformation area is significantly reduced; Figure 4 (b) The structural components of the traditional uniform allowance process have concentrated deformation and large gradient. The present invention has excellent processing deformation control effect.

[0068] This invention also discloses the application of an optimization method for controlling deformation during machining of non-uniform allowance in structural components. The above-mentioned optimization method for controlling deformation during machining of non-uniform allowance in structural components is applied to the control of deformation during the finishing of aerospace structural components, die-forged structural components, or thin-walled precision structural components.

[0069] Therefore, this invention employs the aforementioned optimization method and application for controlling the machining deformation of non-uniform allowances in structural components. By transforming the optimization of non-uniform allowances in structural components into a material distribution optimization problem, and combining the residual stress field and structural stiffness distribution characteristics, the adjoint variable method is used to accurately calculate the sensitivity of design variables and perform spatial filtering. Relying on a two-way progressive optimization strategy to dynamically update the allowance distribution, it can achieve proactive and precise control of machining deformation of structural components. This invention can effectively balance machining deformation control effect and engineering manufacturability, solving the technical problem of difficulty in accurately controlling machining deformation of large and complex structural components due to residual stress release and uneven stiffness distribution. It is applicable to precision machining scenarios of aerospace structural components, die-forged structural components, and thin-walled precision structural components, and can significantly improve the machining accuracy, assembly quality, and service reliability of structural components.

[0070] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method for optimizing a structural part non-uniform allowance machining deformation control process, characterized by the steps of include: S1. Obtain the initial geometric model, material property information and clamping boundary conditions of the structural component before optimization, establish the design domain of the optimization margin, discretize the design domain of the optimization margin into several material layers along the machining depth direction, and define design variables. The design variables are the retention or removal status of each material layer; S2. Based on the initial geometric model, material property information and clamping boundary conditions, a processing deformation evaluation model considering the evolution of structural stiffness during material removal is established. The processing deformation objective function is constructed with the minimum processing deformation of the structural component as the optimization objective, and constraints that meet the upper and lower limits of processing allowance and the requirements of manufacturability are constructed. S3. Based on the processing deformation evaluation model, the sensitivity of each design variable to the processing deformation target is calculated using the adjoint variable method; S4. Perform spatial filtering on the sensitivity calculated in step S3 to obtain the filtered sensitivity. S5. Based on the filtered sensitivity results, a two-way progressive optimization strategy is adopted to delete, retain, or add and update the design variables to obtain a new allowance distribution state. The machining deformation of the structural parts under the new allowance distribution state is calculated. Among them, the machining allowance is reduced first in areas with low sensitivity and strong structural rigidity, and the machining allowance is retained first in areas with high sensitivity and easy structural deformation. S6. Repeat steps S3 to S5 until the preset convergence condition is met to obtain the final non-uniform allowance machining scheme for the structural component.

2. The method of claim 1, wherein: The sensitivity in step S3 is calculated using the adjoint variable method, specifically including: Under the current surplus distribution state, establish the static equilibrium equations for the structural components: ； wherein, is the global stiffness matrix of the structure, is the nodal displacement vector, is the load vector, is the residual distribution state identifier corresponding to the current iteration step; The objective function of machining deformation is expressed as a function of design variables. An adjoint variable with the same dimension as the displacement vector is introduced to construct a Lagrangian function containing static equilibrium constraints. Take the total differential of the Lagrange function, set the partial derivative of the Lagrange function with respect to the displacement variable to zero and satisfy the stationary value condition to obtain the adjoint equation, and solve the adjoint equation to obtain the adjoint variable; Based on the obtained accompanying variables, the sensitivity corresponding to each design variable is calculated. The sensitivity is used to characterize the degree of influence of the corresponding material layer change on the processing deformation target. wherein the first design variable corresponds to the sensitivity, while the second design variable corresponds to the change in the local stiffness matrix and the change in the equivalent nodal force of the residual stress caused by the change in the state of the first design variable.

3. The method of claim 1, wherein: Step S4 performs spatial filtering on the sensitivity of each design variable calculated in step S3. The spatial filtering process obtains the filtered sensitivity by comprehensively considering the sensitivity of adjacent design variables within a preset filtering radius and the distance weight between adjacent design variables and the current design variable.

4. The method of claim 1, wherein: The iterative update of design variables in step S5 adopts a two-way incremental optimization strategy, including: Update the design variables that are currently in the retained state and whose filtered sensitivity is lower than the deletion threshold to the removal state; Update the design variables that are currently removed and whose filtered sensitivity is higher than the added threshold to the retained state; The deletion threshold and the addition threshold are dynamically adjusted based on the sensitivity distribution characteristics and the changing trend of the objective function in the current iteration.

5. The method for optimizing the machining deformation control process of non-uniform allowance in structural components according to claim 1, characterized in that, The final non-uniform machining allowance scheme for the structural component obtained in step S6 satisfies the following allowance allocation rules: For areas with low sensitivity, high structural rigidity, or weak influence from processing deformation, priority should be given to reducing processing allowance; For areas with high sensitivity, key geometric features, weak structures, or easily deformable structures, larger machining allowances should be reserved first. During the design variable update process in step S5, the processing sequence constraint, minimum processing allowance constraint, maximum processing allowance constraint, and manufacturability constraint are satisfied.

6. The method for optimizing the machining deformation control process of non-uniform allowance in structural components according to claim 1, characterized in that, The convergence condition in step S6 includes at least one of the following: the state of the design variables remains unchanged during a preset number of iterations, or the relative change of the processing deformation objective function obtained in two adjacent iterations is less than a preset tolerance.

7. The method for optimizing the machining deformation control process of non-uniform allowance in structural components according to claim 1, characterized in that: In step S1, when the design domain of the margin to be optimized is discretized into several material layers along the processing depth direction, the design variables corresponding to one or more adjacent material layers in the local area of ​​the structural component are merged into a single combined design variable; the single combined design variable is used to characterize the synchronous retention state or synchronous removal state of all material layers in the corresponding range.

8. The method for optimizing the machining deformation control process of non-uniform allowance in structural components according to claim 1, characterized in that, In step S1, the material property information includes basic mechanical property parameters and plasticity parameters; When the material property information also includes the residual stress state, the method for obtaining the residual stress field of the structural component includes at least one of the following: The overall residual stress field of the structural component is directly measured using stress measurement methods. During the pre-processing of the blank, deformation force data reflecting the overall deformation trend of the blank are monitored. A residual stress field inference method based on deformation force is used to infer the residual stress field of the initial blank of the structural component and obtain the initial residual stress field distribution of the structural component.

9. The method for optimizing the machining deformation control process of non-uniform allowance in structural components according to claim 1, characterized in that: In step S2, the processing deformation evaluation model adopts the finite element prediction method or the neural network prediction method. The processing deformation evaluation model is used to simulate the residual stress release and structural deformation during the material removal process under different allowance distribution states. The clamping boundary conditions obtained in step S1 are set according to the actual machining and clamping state, and fixed displacement constraints are applied to the fixed position of the structural component.

10. An application of a process optimization method for controlling deformation during non-uniform allowance machining of structural components, characterized in that, The process optimization method for controlling deformation during non-uniform allowance machining of structural components according to any one of claims 1-9 is applied to the control of deformation during the finishing of aerospace structural components, die-forged structural components, or thin-walled precision structural components.

Images