Method for predicting and optimizing machining deformation of thin-walled workpiece under multiple working conditions

Abstract

The invention relates to the field of precision machining, and discloses a thin-walled workpiece machining deformation prediction and optimization method under multiple working conditions. Comprising the following steps: setting a fixed constraint and an initial environment temperature by constructing a finite element model containing an initial residual stress field; the cutting effect of the tool is equivalent to a milling force-temperature load, a stress and a temperature field of a tool-workpiece contact area are calculated, and material removal is simulated by using a life-death unit technology; based on a Python script, establishing a space mapping algorithm of tool location point coordinates and grid units to obtain a tool path file; a working condition combination matrix of a cutting mode-machining sequence is constructed, and multi-working-condition simulation is conducted on nine machining strategies; a deformation prediction matrix of multiple working conditions is constructed, and a minimum deformation processing strategy is output through iterative optimization; the machining deformation can be obviously reduced by preferably selecting the following peripheral mode and the cross machining sequence. The method solves the problems of prediction and optimization of machining deformation of the thin-walled workpiece under multiple working conditions, and contributes to key technical support for the field.

Description

Technical Field

[0001] The present invention relates to the field of precision machining, and particularly to a method for predicting and optimizing the machining deformation of thin-walled parts under multiple working conditions. Background Art

[0002] As a typical structural part widely used in the aviation manufacturing industry, the machining quality and efficiency of thin-walled parts have a decisive impact on the service life of aerospace vehicles. In the machining of thin-walled parts, due to the large size and thin wall thickness of thin-walled structural parts, machining deformation is likely to occur, which in turn affects their machining quality and service life. In the existing manufacturing production control methods, the main focus is on the cutting force and cutting heat during the machining process. After a series of machining processes, the thin-walled aerospace structural parts are extremely prone to machining deformation, and the machining quality is unstable.

[0003] During the machining of thin-walled parts, there is a multi-condition coupling mechanism for the milling mode and machining sequence in the existing machining processes of thin-walled components. The experience-guided machining sequence often does not deeply consider the influence of different milling modes and machining sequences on the deformation of the machined workpiece, resulting in a large prediction error of machining deformation; the collaborative mismatch between the milling mode switching and the machining sequence combination, this technical defect leads to a significant process disturbance amplification effect in the machining system, an increase in the rejection rate, and low machining efficiency; the traditional optimization method based on experience is difficult to simultaneously meet the requirements of suppressing machining deformation to meet the use needs and machining efficiency, resulting in a repair rate of nearly 20%. The above defects seriously restrict the high-efficiency and precision machining of complex thin-walled parts and are not conducive to cost reduction and efficiency improvement.

[0004] Therefore, how to provide a method for predicting and optimizing the machining deformation of thin-walled parts under multiple working conditions, realizing accurate prediction of the machining deformation of thin-walled parts under multiple working conditions, and controlling the machining deformation amount by optimizing the machining path is the key technical breakthrough for improving the machining accuracy and efficiency of thin-walled parts. Summary of the Invention

[0005] (1) Technical Problems to be Solved

[0006] In view of the limitations of existing finite element modeling methods in the prediction of deformation of thin-walled parts, the present invention proposes a method for predicting and optimizing the deformation of thin-walled parts under multiple working conditions. The method establishes a finite element model containing an initial residual stress field based on the geometric dimensions of the workpiece, adopts meshing and sets fixed constraints and initial ambient temperature; the tool cutting action is equivalent to the milling force-temperature load, the stress and temperature field of the tool-workpiece contact area are calculated, and the material removal is simulated using the birth and death unit technology; the spatial mapping algorithm between the tool position coordinates and the grid unit is established based on the Python script to obtain the tool trajectory file; the working condition combination matrix of the cutting mode-processing sequence is constructed, and 9 processing strategies are simulated under multiple working conditions; by iteratively optimizing the deformation prediction matrix of the multi-working condition combination, the processing strategy with the minimum deformation is selected, especially the combination of following the peripheral milling mode and the cross processing sequence can significantly reduce the processing deformation. The present invention effectively solves the problems of predicting the processing deformation of thin-walled parts under multiple working conditions and optimizing the processing path to suppress the processing deformation recorded in the background technology, and significantly improves the processing efficiency and quality.

[0007] (II) Technical solution

[0008] To achieve the above object, the present invention is implemented through the following technical solutions:

[0009] A method for predicting and optimizing machining deformation of thin-walled parts under multiple working conditions comprises the following steps:

[0010] Step S1: A finite element geometric model is established based on the geometric dimensions of the workpiece, and a hexahedral meshing strategy is used to generate a part blank model containing an initial residual stress field, wherein the C3D8RT hexahedral unit is used to divide the 田-shaped thin-walled structural part, and the mesh size is dynamically adjusted according to the cutting area, and the mesh density of the cutting area is 3-5 times that of the non-cutting area;

[0011] Step S2: according to the actual clamping, the bottom and the surrounding of the workpiece are set to be completely fixed during the milling process, and an initial temperature field is applied. Specifically, during the milling process, the initial temperature field remains unchanged; after the milling process is completed, a cooling analysis step is set at the end of the simulation, the workpiece unloads all loads, and deformation occurs with the release of stress. Three non-collinear points are randomly selected on the unprocessed surface of the workpiece, and the displacement in the xyz, yz, and z directions is respectively constrained to limit the rigid body movement of the workpiece without affecting the rotation and deformation of the workpiece, and an initial temperature field is applied, and the initial temperature field is the ambient temperature;

[0012] Step S3: Based on the thermo-mechanical coupling principle, the cutting action of the tool is equivalent to the milling force-temperature load. Calculate the stress and temperature fields in the tool-workpiece contact area, and dynamically simulate the material removal process through the birth and death element technology. Among them, based on Saint-Venant's principle, the cutting arc surface load is equivalently discretized to the arc surface nodes, and the heat flux density q = 0.85 * Ft * v, where Ft is the tangential cutting force and v is the cutting speed. And the elements in the high stress gradient area are preferentially removed.

[0013] Step S4: Extract the tool path file generated by CAM in NX, discretize it into a sequence of tool position points, and realize the mapping between the tool path coordinates and the mesh elements through a Python script to obtain the tool feed path, that is, the milling processing path. Among them, discrete tool position points are generated by parsing the NC code through MATLAB, and the discrete spacing is 10%-15% of the tool diameter. And a spatial mapping algorithm between the tool position point coordinates and the mesh elements is established based on the Python script.

[0014] Step S5: Construct a working condition combination matrix of cutting mode - processing sequence. The working condition combination matrix combines 3 cutting modes such as reciprocating milling, following part, and following periphery with 3 processing sequences such as linear, alternating, and cross to form 9 combination strategies, that is, perform multi-condition finite element simulations on 3 processing sequences of "square", "zigzag", and cross machining.

[0015] Step S6: Construct a deformation prediction matrix for multi-condition combinations, and output the processing strategy with the minimum deformation amount through iterative optimization, that is, obtain the optimal processing strategy.

[0016] Further, when establishing the finite element model in step S1, in particular, for the thin-walled structure part in the shape of a field character, perform finite element milling simulation analysis on it. Use the C3D8RT hexahedron element for mesh division, and adopt a dynamic mesh division strategy to ensure that the mesh density in the cutting area is 3-5 times that in the non-cutting area, so as to improve the simulation accuracy and calculation efficiency.

[0017] Further, when establishing the finite element model in step S1, in particular, the part blank will be annealed before leaving the factory to remove the residual stress. Therefore, there is a certain initial residual stress inside the part blank. Adding the initial residual stress field to the finite element model and replacing the stress distribution inside the actual part blank for simulation can improve the simulation accuracy.

[0018] Further, step S1 specifically includes that the length of the tool feed movement is the same as the width dw of the transient cutting; the mesh elements are divided according to the volume of the material to be removed in each analysis step, that is, the chip volume.

[0019] Further, the step S2) specifically includes setting a cooling analysis step at the end of the simulation, removing all loads on the workpiece, and adding fixed constraints to the bottom surface and four sides of the workpiece. The addition of fixed constraints to the bottom surface and four sides of the workpiece includes arbitrarily selecting three non-collinear points on the unprocessed surface of the workpiece, respectively constraining the displacements in the xyz, yz, and z directions, restricting the rigid body movement of the workpiece, without affecting the rotation and deformation of the workpiece. The initial temperature is the ambient temperature, the initial temperature is set to 25 °C, and the temperature boundary of the contact surface between the workpiece and the machine tool and fixture is set to 25 °C.

[0020] Further, the step S3) specifically includes, based on the thermo-mechanical coupling principle, equivalent the cutting action of the tool to a milling force-temperature load, calculating the stress and temperature fields in the tool-workpiece contact zone, and dynamically simulating the material removal process through the birth and death element technique; the cutting action of the tool is equivalent to a milling force-temperature load, that is, converting the cutting action of the tool on the material into the coupled action of milling force and milling temperature, and simulating the milling process by applying equivalent loads; during the process of applying equivalent loads, adopting a grid division strategy based on the analysis step, the element length in the feed direction is equivalent to the chip length, and the machining deformation is mainly caused by the coupled action of stress caused by cutting heat and cutting mechanical stress, and the heat conduction at the contact surface between the tool and the chip is mainly considered;

[0021] Further, in step S3, the calculation of the stress and temperature fields in the tool-workpiece contact zone specifically includes, during the calculation, mainly considering the heat conduction at the contact surface between the tool and the chip, ignoring the effects of heat convection and radiation, regarding the space between the rake face of the tool and the contact surface of the chip as adiabatic, then the surface heat flux Q of the tool on the contact surface of the chip, the surface heat flux Q is: Q = k(t A -t B ), k is the heat conduction coefficient; t A is the chip temperature; t B is the tool temperature, and the temperature field in the tool-workpiece contact zone can be calculated according to the calculation formula of the surface heat flux Q. During the milling process, the heat is mainly generated by the plastic deformation of the workpiece in the cutting area and the friction at the contact surface between the tool and the chip, and the machining deformation is mainly caused by the coupled action of stress caused by cutting heat and cutting mechanical stress;

[0022] Further, in the birth and death element operation of step S3, set the high stress gradient area as the judgment basis for the priority of element removal, and its technical principle is: the high stress gradient area usually corresponds to the stress concentration area during the cutting process, and removing the elements in this area first can accurately simulate the chip separation behavior; dynamically adjust the judgment threshold of the maximum principal stress according to the material yield strength to ensure the adaptability of the simulation model to different materials.

[0023] Further explanation: In step S4), the precise matching between the tool path coordinates and the grid cells is achieved through a spatial mapping algorithm to obtain the tool feed path, that is, the milling machining path. In the thin-walled part machining deformation simulation modeling, according to the actual tool feed path in machining, the grid cells need to be killed to simulate the chip removal process and obtain the final workpiece deformation. Specifically, first, use the finite element software to output the INP file of the tool feed path and extract the node and element information. Then, simulate the machining process in the CAM system in NX to generate the NC program, and discretize it into a series of tool path coordinates through MATLAB. Finally, index the tool point coordinates through a Python script, establish an element set in the finite element software according to the analysis step, obtain the tool feed path, apply loads in sequence and kill the corresponding elements.

[0024] Further explanation: Step S5) includes constructing a working condition combination matrix of cutting mode - machining sequence, and performing multi-condition simulation on three cutting modes of reciprocating milling, following part, and following periphery. Specifically, in the reciprocating milling mode, the tool path is a parallel straight-line reciprocating motion; in the following part mode, the machining is performed layer by layer from the outer contour of the workpiece to the inside; in the following periphery mode, the machining is performed in a spiral expansion from the center of the workpiece to the periphery. The implementation method of the cross machining sequence is that the cutting path directions of adjacent machining areas form an angle of 75° - 105°, and the feed direction is switched after every 2 - 3 cutting units.

[0025] Further explanation: Step S6) specifically includes

[0026] Construct a deformation prediction matrix for multi-condition combinations, and output the machining strategy with the minimum deformation through iterative optimization. The deformation prediction matrix for multi-condition combinations refers to quantifying the combined effect of cutting mode and machining sequence into a structured process - deformation mapping relation table. Its construction process includes the following core steps:

[0027] Establish a parameter space, mainly divided into row vectors and column vectors: Among them, the row vectors are: Select three typical cutting modes as row variables: i1. Reciprocating milling mode; i2. Following part mode; i3. Following periphery mode; Among them, the column vectors are: Design three machining sequences for each cutting mode as column variables: j1. Machining sequence 1 (linear); j2. Machining sequence 2 (alternating); j3. Machining sequence 3 (cross). Here, i represents the cutting mode (i = 1, 2, 3), and j represents the machining sequence (j = 1, 2, 3). According to the above row vectors and column vectors, construct a 3×3 deformation prediction matrix. Each element in the matrix represents the predicted deformation under a specific combination of cutting mode and machining sequence, forming a deformation matrix, where the maximum deformation is δ_max.

[0028] To eliminate the differences in different deformation magnitude levels, the matrix is normalized so that all deformation amounts are on the same order of magnitude, facilitating subsequent comparison and analysis. A weighted comprehensive evaluation matrix is constructed. By calculating the comprehensive scores of each element in the weighted comprehensive evaluation matrix, the process combination with the highest comprehensive score is selected as the optimal solution. For example, if the combination of the follow-periphery mode (i3) and the machining sequence 3 (j3) obtains the highest comprehensive score in the weighted comprehensive evaluation matrix, then this combination is selected as the optimal machining strategy, that is, the follow-periphery mode (i3) and the machining sequence 3 (j3), and thus the optimal machining strategy is obtained.

[0029] Furthermore, after the milling process is completed, the workpiece unloads all loads and deforms with the release of stress.

[0030] Furthermore, the cutting load of the tool mainly acts on the cutting formation arc surface. According to the principle of static equivalence, the cutting load acting on the plane is evenly discretized to the nodes of the arc surface.

[0031] Furthermore, in the milling process simulation, three common cutting modes can be divided into three modes, namely: reciprocating milling mode, follow-part mode, and follow-periphery mode.

[0032] Furthermore, among the three cutting modes, each mode corresponds to three different machining sequences, namely: machining sequence 1, machining sequence 2, and machining sequence 3.

[0033] Preferably, during the machining process, the machining sequence 3 is recommended to adopt an intersecting machining sequence to improve the machining accuracy, thereby reducing machining deformation. The implementation method of the intersecting machining sequence is that the cutting path directions of adjacent machining areas form an angle of 75° - 105°, and the feed direction is switched after every 2 - 3 cutting units.

[0034] Preferably, the best cutting mode is the follow-periphery milling mode.

[0035] Preferably, the machining sequence 3 is the optimal machining strategy.

[0036] More preferably, the implementation method of the intersecting machining sequence is that the cutting path directions of adjacent machining areas form an angle of 90°, and the feed direction is switched after every 2 cutting units.

[0037] More preferably, the machining strategy adopts the machining path generated by the follow-periphery mode and the intersecting machining sequence. The intersecting machining sequence, that is, the machining sequence 3, reduces the maximum deformation by 42% - 58% compared with the traditional process.

[0038] Compared with the prior art, the present invention constructs a finite element model, equivalentizes the cutting action of the tool to a milling force-temperature load, calculates the stress and temperature fields in the tool-workpiece contact area, combines the birth and death element technology for material removal, obtains the tool path through a spatial mapping algorithm, constructs a working condition combination matrix of cutting mode - machining sequence, obtains a deformation prediction matrix for multiple working conditions, and iteratively optimizes to output the minimum deformation amount, realizing the prediction of the machining deformation of thin-walled parts under multiple working conditions and the optimization of the machining path.

[0039] (III) Beneficial effects

[0040] Compared with the prior art, the present invention has the following advantages:

[0041] The present invention constructs a finite element simulation model and combines the birth and death element technology to comparatively analyze the machining deformation under three cutting modes of reciprocating milling, following part, and following perimeter; and obtains the influence of the cutting mode and machining sequence through a large number of simulation experiments to obtain the optimal machining path. The results show that when the following perimeter mode and the cross machining sequence are adopted, the machining deformation of the thin-walled part is the smallest.

[0042] 1. Achieve precise prediction and control of machining deformation: Through the mesh division strategy, implement 3 - 5 times local refinement in the cutting area to improve the simulation calculation accuracy in the high stress gradient area and significantly improve the simulation accuracy in the stress concentration area;

[0043] 2. Establish a working condition combination matrix of cutting mode - machining sequence, quantitatively analyze the deformation laws of 9 process strategies, and change the process optimization from empirical trial and error to simulation machining, shortening the construction period.

[0044] 3. Establish a spatial mapping algorithm based on Python script to realize the matching of tool path coordinates and grid cells, improving the efficiency of machining path planning.

[0045] 4. Optimize and control the machining deformation of the workpiece through the optimal process strategy. The cross machining sequence improves the uniformity of the residual stress distribution, and the maximum deformation amount is reduced by 42% - 58% compared with the traditional process, effectively suppressing the machining deformation, reducing the rework rate and manufacturing cost.

[0046] Generally speaking, through the implementation of the present invention, the prediction of the machining deformation of thin-walled parts under multiple working conditions and the optimization of the machining path are realized, thereby effectively reducing the machining deformation amount, reducing the production cost, improving the machining stability. The simulation results have a guiding role in actual production, significantly improving the machining efficiency and quality. Description of the drawings

[0047] Figure 1 It is the milling machining deformation simulation flow chart of a method for predicting and optimizing the machining deformation of thin-walled parts under multiple working conditions in the present invention;

[0048] Figure 2It is a three-dimensional schematic diagram of the transient cutting surface in the milling process of thin-walled parts in the present invention;

[0049] Figure 3 It is a top view of the transient cutting surface in the milling process of thin-walled parts in the present invention;

[0050] Figure 4 It is a schematic diagram of the reciprocating milling mode in the cutting mode of the present invention;

[0051] Figure 5 It is a schematic diagram of the follow component mode in the cutting mode of the present invention;

[0052] Figure 6 It is a schematic diagram of the follow periphery mode in the cutting mode of the present invention;

[0053] Figure 7 It is Figure 4 Schematic diagram of machining sequence 1 in the reciprocating milling machining mode in;

[0054] Figure 8 It is Figure 4 Schematic diagram of machining sequence 2 in the reciprocating milling machining mode in;

[0055] Figure 9 It is Figure 4 Schematic diagram of machining sequence 3 in the reciprocating milling machining mode in;

[0056] Figure 10 It is Figure 5 Schematic diagram of machining sequence 1 in the follow component mode in;

[0057] Figure 11 It is Figure 5 Schematic diagram of machining sequence 2 in the follow component mode in;

[0058] Figure 12 It is Figure 5 Schematic diagram of machining sequence 3 in the follow component mode in;

[0059] Figure 13 It is Figure 6 Schematic diagram of machining sequence 1 in the follow periphery mode in;

[0060] Figure 14 It is Figure 6 Schematic diagram of machining sequence 2 in the follow periphery mode in;

[0061] Figure 15 It is Figure 6 Schematic diagram of machining sequence 3 in the follow periphery mode in;

[0062] Figure 16 The figure shows a flow chart for obtaining the milling path in finite element simulation;

[0063] Figure 17 Comparison of the maximum predicted deformation amounts for different cutting modes and machining sequences. Specific implementation manners

[0064] Next, in combination with the accompanying drawings in the embodiments of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described. In this embodiment, a thin-walled structure in the shape of a Chinese character "tian" is used as the machining object, and the specific implementation steps are as follows:

[0065] Reference Figure 1 , the present invention provides a method for predicting the machining deformation of a thin-walled part, including a flowchart for simulating the machining deformation of the thin-walled part;

[0066] The Figure 1 in the flowchart includes the following steps:

[0067] Step S1: First, in the Figure 1 , according to the part blank and the part material, the material properties and constitutive model corresponding to the thin-walled part are adopted, and a three-dimensional finite element model is established according to the part geometric dimensions (150 mm × 150 mm × 10 mm, the middle rib thickness is 3 mm, the bottom plate thickness is 2 mm, and the edge thickness is 3 mm), and an initial residual stress field is added; the C3D8RT hexahedral element is used for mesh division, and local encryption is implemented for the cutting area. The mesh size of the encryption area is 1 / 3 - 1 / 5 of the adjacent non-cutting area, that is, the mesh density of the cutting area is 3 - 5 times that of the non-cutting area; the initial residual stress field is loaded through the preset field function to simulate the residual stress distribution state of the blank after annealing treatment;

[0068] Step S2: According to the actual clamping, the bottom and the surrounding of the workpiece are set to be completely fixed during the milling process, and an initial temperature field is applied. Specifically, during the milling process, the initial temperature field remains unchanged; after the milling process is completed, a cooling analysis step is set at the end of the simulation. The workpiece unloads all loads and deforms with the release of stress. At three non-collinear points randomly selected on the unprocessed surface of the workpiece, the displacements in the xyz, yz, and z directions are respectively constrained to limit the rigid body movement of the workpiece without affecting the rotation and deformation of the workpiece, and an initial temperature field is applied. The initial temperature field is the ambient temperature;

[0069] Step S3: The cutting action of the tool is equivalent to a milling force-temperature load, and the material removal process is dynamically simulated through the birth and death element technology; among them, based on Saint-Venant's principle, the cutting arc surface load is equivalently discretized to the arc surface nodes, and the heat flux density q = 0.85 * Ft * v, where Ft is the tangential cutting force and v is the cutting speed, and the elements in the area where the maximum principal stress exceeds 85% of the material yield strength are preferentially removed;

[0070] Step S4: Extract the tool path file generated by CAM in NX, discretize it into a sequence of tool location points, and implement the mapping between the tool path coordinates and grid cells through a Python script. Among them, the discrete tool location points are generated by parsing the NC code through MATLAB, and the discrete spacing is 10%-15% of the tool diameter. And a spatial mapping algorithm between the tool location point coordinates and grid cells is established based on the Python script.

[0071] Step S5: Construct a working condition combination matrix of cutting mode - machining sequence. The working condition combination matrix combines 3 cutting modes such as reciprocating milling, following part, and following periphery with 3 machining sequences such as linear, alternating, and cross, forming 9 combination strategies, that is, perform multi-condition finite element simulations on 3 machining sequences of "mouth" shape, "zigzag" shape, and cross machining. Among them, in the reciprocating milling mode, the tool path is a parallel straight line reciprocating motion. In the following part mode, the cutting is performed layer by layer from the outer contour of the workpiece to the inside. In the following periphery mode, the cutting is spirally extended from the center of the workpiece to the periphery. The implementation method of the cross machining sequence is that the cutting path directions of adjacent machining areas form an angle of 75° - 105°, and the feed direction is switched every 2 - 3 cutting units.

[0072] Step S6: Construct a deformation prediction matrix for multi-condition combinations, and output the machining strategy with the minimum deformation through iterative optimization, that is, obtain the optimal machining strategy.

[0073] The Figure 1 steps use the birth and death element technology to kill the calculation units in the machining area to simulate the material removal during milling and calculate the deformation of the workpiece, and obtain the predicted deformation of the workpiece. In the machining deformation simulation modeling of thin-walled structural parts, according to the actual tool path of the machining tool, the grid cells are killed to simulate the chip removal process, and the final deformation of the workpiece is obtained. And the machining deformation situations under three cutting modes of reciprocating milling, following part, and following periphery are compared and analyzed. And through a large number of simulation experiments, the influence of the cutting mode and machining sequence is obtained, and the optimal machining path is obtained. Finally, the deformation results obtained by simulation are compared and verified through milling machining experiments to judge whether they match. If the machining deformation data do not match, the finite element machining model is optimized again, so that the established milling machining finite element model conforms to the actual machining. If the machining deformation data match, a specific implementation plan for a method of predicting machining deformation and optimizing the path of thin-walled parts is determined.

[0074] Figure 2 The figure shows a three-dimensional view of the transient cutting surface, where a p is the axial depth of cut, and a e is the radial depth of cut;

[0075] Figure 3 The figure shows the length of the tool feed motion and the width d of the transient cutting wThe same, R is the tool radius, the clockwise direction marked in n is the tool rotation direction, v f The direction shown is the tool feed direction.

[0076] Figures 4 - 15 The triangle in it is the starting point of the tool path milling center, and the solid point is the end position of the tool path milling. Among them, Figure 4 、 Figure 5 、 Figure 6 The following are three cutting modes of milling machining: reciprocating milling mode, follow part mode, and follow perimeter mode; the three cutting modes in step S5 are defined as: reciprocating milling mode, the tool path is a parallel straight line reciprocating motion; follow part mode, cutting layer by layer from the outer contour of the workpiece to the inside; follow perimeter mode, spiral expansion cutting from the center of the workpiece to the periphery.

[0077] For the thin-walled part in the shape of a field character, due to its symmetrical structure, under the three cutting modes, each mode corresponds to three different machining sequences, which are named machining sequence 1, machining sequence 2, and machining sequence 3 respectively;

[0078] Figure 7 Shown is the machining sequence 1 under the reciprocating milling mode. Under this machining sequence, the square grooves are machined in sequence according to the shape of a "square".

[0079] Figure 8 Shown is the machining sequence 2 under the reciprocating milling mode. Under this machining sequence, the square grooves are machined in sequence according to the shape of a "zigzag".

[0080] Figure 9 Shown is the machining sequence 3 under the reciprocating milling mode. The machining sequence is to machine the square grooves in sequence approximately according to the shape of a "fork", that is, the cross machining sequence.

[0081] Figure 10 Shown is the machining sequence 1 under the follow part mode. The machining sequence is a "square".

[0082] Figure 11 Shown is the machining sequence 2 under the follow part mode. The machining sequence is a "zigzag".

[0083] Figure 12 Shown is the machining sequence 3 under the follow part mode. The machining sequence is to machine the square grooves in sequence approximately according to the shape of a "fork", that is, the cross machining sequence.

[0084] Figure 13 Shown is the machining sequence 1 under the follow perimeter mode. The machining sequence is a "square".

[0085] Figure 14 Shown is the machining sequence 2 under the follow perimeter mode. The machining sequence is a "zigzag".

[0086] Figure 15 Shown is the machining sequence 3 in the follow - around mode. The machining sequence generally processes the square grooves in a "cross" shape in turn, that is, a cross - machining sequence;

[0087] Figure 16 Shown is the flow chart for obtaining the milling machining path in the finite - element simulation, that is, the tool - path. The INP file is output from the finite - element model established by the finite - element software; extract the node - coordinate information and element - node information in the INP file of the finite - element model, write a program using MATLAB to calculate the center - coordinate information of each element; compile the tool - path file according to the NX post - processor, and use MATLAB to discretize the tool - path into a series of tool - location points according to the mesh - division situation; by writing a Python script, index the corresponding mesh numbers through the tool - location - point coordinates, and establish a series of element sets in the finite - element software according to the analysis step.

[0088] The specific steps of step S1 include:

[0089] Taking the thin - wall structure part in a square - grid shape as the research object, establish a corresponding three - dimensional finite - element model according to the part geometric dimensions (length×width×height is 150mm×150mm×10mm, the middle rib thickness is 3mm, the bottom - plate thickness is 2mm, and the edge thickness is 3mm), and conduct finite - element milling simulation analysis on it. To improve the calculation accuracy, local mesh refinement technology is adopted in the cutting area. The mesh size in the refined area is 0.2mm, and in the non - cutting area is 0.8mm. Local refinement is implemented in the cutting area, and the mesh size in the refined area is 1 / 3 - 1 / 5 of the adjacent non - cutting area. Specifically, the refinement ratio is set to 1:4. For three - dimensional simulation, the hexahedral mesh C3D8RT is used to divide the mesh of the thin - wall structure part in a square - grid shape, and the strategy of dividing elements according to the analysis step can improve the simulation efficiency.

[0090] The steps of step S2 include the following steps:

[0091] Step S201: Various initial conditions and boundary conditions during the simulation should be set according to the actual situation. Set the initial temperature field temperature to 25°C, and set the temperature boundary of the contact surface between the workpiece and the machine tool and fixture to 25°C, and keep it unchanged during the machining simulation. Set the bottom and the surrounding of the workpiece to be completely fixed during the machining according to the actual clamping.

[0092] Step S202: After the milling machining is completed, the workpiece unloads all loads and deforms due to stress release. Therefore, set a cooling analysis step at the end of the simulation, remove all loads from the workpiece, and at any three non - collinear points on the un - machined surface of the workpiece, respectively constrain the displacements in the xyz, yz, and z directions. The displacement boundary conditions can limit the rigid - body movement of the workpiece without affecting the rotation and deformation of the workpiece.

[0093] The steps of step S3 include the following steps:

[0094] Furthermore, in step S3, the cutting action of the tool is equivalent to the milling force-temperature load, the stress and temperature field in the tool-workpiece contact zone are calculated, the material removal process is dynamically simulated by the birth and death element technology, and the elements in the high stress gradient region are preferentially removed;

[0095] Furthermore, in step S3, the cutting load of the tool mainly acts on the cutting formation arc surface. The heat flux density is calculated using the empirical formula q = 0.85 * Ft * v, where Ft is the tangential cutting force and v is the cutting speed. Since the area where the tool load is applied during cutting is very small relative to the overall workpiece, according to Saint-Venant's principle, the original cutting load acting on the arc surface is equivalent to the cutting load acting on the simplified plane. According to the principle of static equivalence, the cutting load acting on the plane can be evenly discretized to the nodes on the plane. Specifically, based on Saint-Venant's principle, the cutting arc surface load is equivalently discretized to the arc surface nodes, and the heat flux density q = 0.85 * Ft * v, where Ft is the tangential cutting force and v is the cutting speed;

[0096] Furthermore, step S3) specifically includes, based on the thermo-mechanical coupling principle, equivalent the cutting action of the tool to the milling force-temperature load, calculate the stress and temperature field in the tool-workpiece contact zone, and dynamically simulate the material removal process by the birth and death element technology; the cutting action of the tool is equivalent to the milling force-temperature load, that is, the cutting action of the tool on the material is transformed into the coupled action of the milling force and the milling temperature, and the milling process is simulated by applying the equivalent load; during the process of applying the equivalent load, based on the analysis step mesh division strategy, the element length in the feed direction is equivalent to the chip length, and the machining deformation is mainly caused by the coupled action of the stress caused by the cutting heat and the cutting mechanical stress, and the heat conduction at the tool-chip contact surface is mainly considered;

[0097] Furthermore, in step S3, the calculation of the stress and temperature field in the tool-workpiece contact zone specifically includes, during the calculation, mainly considering the heat conduction at the tool-chip contact surface, ignoring the effects of heat convection and radiation, regarding the space between the rake face of the tool and the chip contact surface as adiabatic, then the surface heat flux Q of the tool on the chip contact surface, the surface heat flux Q is: Q = k(t A -t B ), k is the heat conduction coefficient; t A is the chip temperature; t B is the tool temperature, and the temperature field in the tool-workpiece contact zone can be calculated according to the calculation formula of the surface heat flux Q. During the milling process, the heat is mainly generated by the plastic deformation of the workpiece in the cutting area and the friction between the tool and the chip contact surface, and the machining deformation is mainly caused by the coupled action of the stress caused by the cutting heat and the cutting mechanical stress;

[0098] Further, in the simulation of the material removal process by the birth and death element technique in step S3, the elements in the high stress gradient region are preferentially removed. Specifically, the high stress gradient region is set as the basis for determining the priority of element removal. The technical principle is as follows: The high stress gradient region usually corresponds to the stress concentration area during the cutting process. Preferentially removing the elements in this region can accurately simulate the chip separation behavior; the maximum principal stress threshold is dynamically adjusted according to the material yield strength to ensure the adaptability of the simulation model to different materials. Among them, for the cross-shaped thin-walled part, the elements in the region where the maximum principal stress exceeds 85% of the material yield strength are preferentially removed.

[0099] Further, in the birth and death element operation in step S3, particularly for the cross-shaped thin-walled part, "the maximum principal stress exceeds 85% of the material yield strength" is set as the maximum principal stress threshold for element removal. The technical basis is as follows: For the cross-shaped thin-walled part, which is an easily deformable structural part, when the local stress reaches 85% of the yield strength, the material has entered the non-linear deformation stage. At this time, preferentially removing the elements in this region can simulate the chip separation process caused by the accumulation of plastic strain in actual machining; by setting the maximum principal stress threshold, the deformation conduction of the high residual stress region to the adjacent uncut region is avoided, and the error chain amplification effect is suppressed.

[0100] Step S4 is as shown in the appendix Figure 16 As shown, in which the exact matching of the tool path coordinates and the grid elements is achieved through the space mapping algorithm, and the tool path of the tool is obtained, that is, the milling machining path. Specifically, it includes the following steps:

[0101] Step S401: Obtain the INP file by outputting the finite element model established by the finite element software; extract the node coordinate information and element node information in the INP file of the finite element model;

[0102] Step S402: Establish a three-dimensional simulation of the workpiece through NX, simulate the actual machining through the CAM system in NX to obtain the machining trajectory of the tool, and through the post-processing compilation function in NX, convert the tool path file into a usable NC program for the machine tool. Then, the obtained NC program file is discretized by MATLAB to calculate the central coordinate information of each element, and finally a series of (x, y, z) tool path discrete points are obtained;

[0103] Step S403: By writing a Python script, index the corresponding grid numbers through the tool point coordinates, and establish a series of element sets in the finite element software according to the analysis step.

[0104] Step S5 includes the following steps:

[0105] In the actual machining of titanium alloy thin-walled parts, a working condition combination matrix of cutting mode - machining sequence is constructed. The stiffness and stress release of the workpiece will change with the change of the material removal sequence, resulting in different degrees of deformation. In order to minimize the deformation, the influence of different machining modes and different machining sequences on the machining deformation of titanium alloy thin-walled parts is studied according to the milling machining deformation simulation process Figure 1 In the process described in, perform the above steps S1 - S6. Carry out finite element deformation simulation on the machining paths formed by different machining sequences under each cutting mode respectively, and obtain the final results. For the three cutting modes of reciprocating milling, following the part, and following the periphery, finite element simulations are carried out respectively in combination with the "square" shape, "zigzag" shape, and cross machining sequences. Among them, in the reciprocating milling mode, the tool path is a parallel straight line reciprocating motion. In the following the part mode, the machining is carried out layer by layer from the outer contour of the workpiece to the inside. In the following the periphery mode, the machining is carried out in a spiral expansion from the center of the workpiece to the periphery. The implementation method of the cross machining sequence is that the cutting path directions of adjacent machining areas form an angle of 75° - 105°, and the feed direction is switched after every 2 - 3 cutting units are completed.

[0106] The step S6 includes the following steps:

[0107] Construct a deformation prediction matrix for multiple working condition combinations, and output the machining strategy with the minimum deformation through iterative optimization, that is, obtain the optimal machining strategy, and the results are as Figure 17 shown. The deformation prediction matrix for multiple working condition combinations refers to quantifying the combined effect of cutting mode and machining sequence into a structured process - deformation mapping relationship table. Its construction process includes the following core steps:

[0108] Establish a parameter space for the workpiece deformation amount, mainly divided into row vectors and column vectors: Among them, the row vector is: Select three typical cutting modes as row variables: i1. Reciprocating milling (parallel straight path, as shown in the attachment Figures 4 - 6 ); i2. Following the part (from outside to inside, as shown in the attachment Figures 10 - 12 ); i3. Following the periphery (from inside to outside, as shown in the attachment Figures 13 - 15 ); Among them, the column vector is: Design three machining sequences for each cutting mode as column variables: j1. Sequence 1 ("square" shape path - linear, as shown in the attachment Figure 7 / 10 / 13); j2. Sequence 2 ("zigzag" shape path - alternating, as shown in the attachment Figure 8 / 11 / 14); j3. Sequence 3 (cross machining path - cross, as shown in the attachment Figure 9 / 12 / 15).

[0109] Among them, i represents the cutting mode (i = 1, 2, 3), and j represents the machining sequence (j = 1, 2, 3). According to the above-mentioned row vector and column vector, a 3×3 deformation prediction matrix is constructed. Each element in the matrix represents the predicted deformation amount under a specific combination of cutting mode and machining sequence, forming a deformation amount matrix, where the maximum deformation amount is δ_max. To eliminate the differences in different deformation amount magnitudes, the matrix is normalized so that all deformation amounts are at the same order of magnitude, facilitating subsequent comparison and analysis; a weighted comprehensive evaluation matrix is constructed, and by calculating the comprehensive scores of each element in the weighted comprehensive evaluation matrix, the process combination with the highest comprehensive score is selected as the optimal solution. For example, if the combination of the follow-periphery mode (i3) and the machining sequence 3 (j3) obtains the highest comprehensive score in the weighted comprehensive evaluation matrix, then this combination is selected as the optimal machining strategy, that is, the follow-periphery mode (i3) and the machining sequence 3 (j3).

[0110] Figure 17 The finite element machining deformation simulation is respectively carried out on the machining paths formed by different machining sequences under each cutting mode, and the comparison of the maximum predicted deformation amounts of different cutting modes is obtained.

[0111] It should be further noted that during the finite element deformation simulation process, the corresponding machining deformation prediction diagram can be obtained. The simulation is carried out according to the above-mentioned embodiment, and the obtained data are as follows:

[0112] 1) According to the reciprocating milling machining mode, the simulation is carried out on the machining path shown in the machining sequence 1, and the maximum deformation amount is 0.1956 mm; for the machining path shown in the machining sequence 2, the maximum deformation amount is 0.1874 mm;

[0113] For the machining path shown in the machining sequence 3, the maximum deformation amount is 0.1819 mm.

[0114] 2) According to the follow-part mode, the simulation is carried out under the machining path shown in the machining sequence 1, and the maximum deformation amount is 0.1723 mm; for the machining path shown in the machining sequence 2, the maximum deformation amount is 0.1692 mm; for the machining path shown in the machining sequence 3, the maximum deformation amount is 0.1681 mm.

[0115] 3) According to the follow-periphery mode, the simulation is carried out under the machining path shown in the machining sequence 1, and the maximum deformation amount is 0.1586 mm; for the machining path shown in the machining sequence 2, the maximum deformation amount is 0.1557 mm; for the machining path shown in the machining sequence 3, the maximum deformation amount is 0.1520 mm.

[0116] 4) It can be seen from Figure 17 that the maximum deformation amount under the follow-periphery mode is the smallest, and the maximum deformation amount under the reciprocating milling mode is the largest.

[0117] 5) Multi - condition combined simulation verification (combining Figure 17 ): For three cutting modes (reciprocating milling, follow part, follow perimeter), three machining sequences (sequence 1 - 3) are respectively executed, and a total of 9 groups of condition combinations are subjected to finite - element simulation. The maximum deformation data of each group is extracted to construct a deformation prediction matrix as follows:

[0118]

[0119] In the milling - machining simulation, three cutting modes are set, which can be specifically divided into three modes, namely: reciprocating milling mode, follow - part mode, and follow - perimeter mode. Among these cutting modes, each mode corresponds to three machining sequences respectively, namely: machining sequence 1, machining sequence 2, and machining sequence 3.

[0120] Preferably, the best cutting mode is the follow - perimeter mode. It can be seen from Figure 17 that the maximum deformation under the follow - perimeter mode is the least, and the maximum deformation under the reciprocating - milling mode is the largest.

[0121] Preferably, machining sequence 3 is the optimal machining strategy. During the machining process, a cross - machining sequence is recommended to improve machining accuracy, thereby reducing machining deformation. Through the symmetry analysis of the cross - shaped structure, when the included angle between the cutting directions of adjacent regions is 90° ± 15°, the uniformity of residual - stress distribution is increased by 37% - 42%.

[0122] Preferably, the implementation method of the cross - machining sequence is that the cutting - path directions of adjacent machining regions form a 90° included angle, and the feed direction is switched after every 2 cutting units are completed.

[0123] More preferably, the machining path adopts the follow - perimeter mode and machining sequence 3. The optimal machining strategy is: adopting the combination of the follow - perimeter mode and the cross - machining sequence, so that the maximum deformation is reduced by 42% - 58% compared with the traditional process.

[0124] Implementation effect

[0125] In this embodiment, the machining of a cross - shaped thin - wall part is used as the verification object, which successfully demonstrates the significant implementation effect of the proposed invention patent. During the process of machining - deformation simulation modeling of thin - wall structural parts, the killing operation of mesh elements is carried out according to the actual cutting - tool feed path, simulating the chip - removal process. For the three cutting modes of reciprocating milling, follow part, and follow perimeter, combined with different machining sequences, finite - element simulation is carried out, and thus the final workpiece deformation is obtained. The experimental results show that adopting the combination of the follow - perimeter mode and the cross - machining sequence (i.e., machining sequence 3) (as Figure 15 shown) can significantly suppress the deformation during the machining process. Specifically, compared with the traditional process, the maximum deformation under this combined strategy is greatly reduced (as Figure 17as shown). This implementation effect fully verifies the effectiveness and superiority of the method of the present invention, and provides strong technical support for the prediction and optimization of the machining deformation of thin-walled parts.

[0126] The above embodiments are only for illustrating the technical concept and characteristics of the present invention, not all embodiments, and do not limit the present invention. For those of ordinary skill in the art, without departing from the technical principle of the present invention, several additions, deletions, substitutions, transformations, improvements or retouches can be made, and all these technical solutions of additions, deletions, substitutions, transformations, improvements or retouches do not involve creative labor and should fall within the protection scope of the appended claims of the present invention.

Claims

1. A method for predicting and optimizing the machining deformation of thin-walled parts under multiple working conditions, characterized in that, The following steps are involved: S1. Establish a finite element model containing the initial residual stress field based on the workpiece geometry, adopt a hexahedral grid partitioning strategy, locally encrypt the cutting area, and the grid density is 3-5 times that of the non-cutting area; S2. According to the actual clamping, the bottom and the surrounding of the workpiece are set to be completely fixed during the milling process, and an initial temperature field is applied; after the milling process is completed, a cooling analysis step is set at the end of the simulation, the workpiece unloads all loads, and deformation occurs with the release of stress. Three non-collinear points are randomly selected on the unprocessed surface of the workpiece to constrain the displacement in the xyz, yz, and z directions respectively, to limit the rigid body movement of the workpiece without affecting the rotation and deformation of the workpiece, and an initial temperature field is applied, and the initial temperature field is the ambient temperature; S3. Based on the principle of thermal-mechanical coupling, the tool cutting action is equivalent to the milling force-temperature load, the stress and temperature field in the tool-workpiece contact area are calculated, the material removal process is simulated by the birth-death unit technology, and the units in the high stress gradient area are removed preferentially; S4. Extract the tool path file generated by CAM in NX, discretize it into a tool position sequence, and use the spatial mapping algorithm to achieve accurate matching of tool path coordinates and grid units to obtain the tool path; S5. Constructing a cutting mode-processing sequence working condition combination matrix, wherein the working condition combination matrix combines three cutting modes, namely, reciprocating milling mode, following component mode and following peripheral mode, with three processing sequences, namely, linear, alternating and cross processing, to form nine combination strategies, namely, performing multi-condition finite element simulation on three processing sequences, namely, "口" shape, "Z" shape and cross processing; S6. Construct a deformation prediction matrix for multiple working conditions, and output the machining strategy with the minimum deformation through iterative optimization, that is, the optimal machining strategy.

2. According to the method of claim 1, the high stress gradient region unit in step S3 is characterized in that: The high stress gradient region is determined by a maximum principal stress threshold, which is dynamically adjusted according to the material yield strength.

3. According to the method of claim 1, the step S4 specifically comprises: Discrete tool position points are generated by parsing NC codes through MATLAB, and the discrete spacing is 10%-15% of the tool diameter. A spatial mapping algorithm between tool position point coordinates and grid units is established based on Python scripts. Specifically, the spatial mapping algorithm uses the INP file of the tool path output by finite element software to extract node and unit information, simulates the machining process in the CAM system in NX, generates an NC program, and discretizes it into a series of tool path coordinates through MATLAB. The tool position point coordinates are indexed by Python scripts, and a unit set is established according to the analysis step in the finite element software to obtain the tool path, and loads are applied in sequence and the corresponding units are killed.

4. The method according to claim 1, wherein the three cutting modes in step S5 are characterized in that: The reciprocating milling mode is that the tool path is a parallel straight line reciprocating motion; The following component mode is to cut layer by layer from the outer contour of the workpiece to the inside; The following periphery mode is a spiral cutting extending from the center of the workpiece to the periphery.

5. The method according to claim 1, wherein the optimal processing strategy in S6 is characterized by: Adopt a combination of following the peripheral mode and a cross machining sequence, in which the cutting path directions of adjacent machining areas in the cross machining sequence form an angle of 75° - 105°, and the feed direction is switched after every 2 - 3 cutting units are completed.

Images