A high-precision tool path planning method and system considering material deformation compensation

Abstract

The application discloses a high-precision tool path planning method and system considering material deformation compensation, and relates to the technical field of numerical control machining path optimization.The method comprises the following steps: carrying out up-down composite detection on a workpiece to be machined, establishing a workpiece size difference grid and performing finite element analysis to obtain the deformation characteristics; constructing a workpiece material characteristic matrix, combining a design scheme to determine a target tool from a tool library; constructing a machining process reinforcement chain to plan an initial tool path; compensating the initial tool path based on the deformation characteristics and obtaining an optimized first result through hierarchical optimization; and obtaining a final optimized tool path through adaptive corner avoidance compensation.The application solves the technical problem that the traditional machining path planning method fails to adapt to key influencing factors in actual machining, resulting in insufficient machining precision, difficult risk control and poor efficiency, and achieves the technical effects of realizing high-precision tool path planning, guaranteeing machining quality, controlling machining risk and improving machining efficiency.

Description

Technical Field

[0001] This invention relates to the field of CNC machining path optimization technology, and in particular to a high-precision tool path planning method and system that takes into account material deformation compensation. Background Technology

[0002] In high-precision machining, toolpath planning directly impacts machining accuracy, efficiency, and tool life, and its scientific nature is crucial for manufacturing high-value-added parts. Existing technologies largely employ traditional path planning methods, which are effective in machining scenarios with simple structures and negligible material deformation. However, as precision manufacturing demands increase, traditional technologies are gradually revealing their limitations. Traditional path planning does not incorporate industrial mechanisms such as materials mechanics and cutting engineering, failing to accurately capture the multi-directional material deformation characteristics of the workpiece. This results in poor adaptability between the path and the actual machining process, easily leading to machining deviations, tool wear, and other problems, making it difficult to meet the precise planning and risk management requirements of high-precision machining. Summary of the Invention

[0003] This application provides a high-precision toolpath planning method and system that considers material deformation compensation, which solves the technical problem that traditional machining path planning methods fail to adapt to key influencing factors in actual machining, resulting in insufficient machining accuracy, uncontrollable risks, and poor efficiency.

[0004] The first aspect of this application provides a high-precision toolpath planning method considering material deformation compensation. The method includes: performing upper and lower composite detection on the workpiece to be machined according to a workpiece design scheme; establishing a workpiece size difference mesh; and performing multi-directional material deformation finite element analysis on the workpiece to be machined based on the workpiece size difference mesh to obtain multi-directional material deformation characteristics; constructing a workpiece material characteristic matrix for the workpiece to be machined; selecting a target tool from a tool library based on the workpiece design scheme; constructing a machining process reinforcement chain; and planning a path for the target tool based on the workpiece design scheme and the machining process reinforcement chain to obtain an initial toolpath; performing material deformation compensation on the initial toolpath based on the multi-directional material deformation characteristics to obtain an adjustable toolpath space; and performing hierarchical optimization of the adjustable toolpath space using a machining risk assessment model to obtain a first result of toolpath optimization; and performing adaptive corner avoidance compensation for the target tool on the first result of toolpath optimization to obtain a second result of toolpath optimization.

[0005] A second aspect of this application provides a high-precision toolpath planning system considering material deformation compensation. The system includes: a material deformation characteristic acquisition module, used to perform upper and lower composite detection on the workpiece to be machined according to the workpiece design scheme, establish a workpiece size difference mesh, and perform multi-directional material deformation finite element analysis on the workpiece to be machined based on the workpiece size difference mesh to obtain multi-directional material deformation characteristics; a target tool acquisition module, used to construct a workpiece material characteristic matrix of the workpiece to be machined, and perform fitness selection on a tool library based on the workpiece design scheme to determine the target tool; an initial toolpath acquisition module, used to construct a machining process reinforcement chain, and perform path planning on the target tool according to the workpiece design scheme and the machining process reinforcement chain to obtain an initial toolpath; a first result acquisition module, used to perform material deformation compensation on the initial toolpath based on the multi-directional material deformation characteristics, obtain an adjustable toolpath space, and perform hierarchical optimization of the adjustable toolpath space through a machining risk assessment model to obtain a first result of toolpath optimization; and a second result acquisition module, used to perform adaptive corner avoidance compensation of the target tool on the first result of toolpath optimization to obtain a second result of toolpath optimization.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] This application achieves precise adaptation of the tool path to the actual machining scenario by performing combined upper and lower inspection and material property analysis on the workpiece to be machined, combining machining requirements to complete tool adaptation and machining process enhancement, and after path planning, incorporating material deformation compensation and machining risk layering optimization, and superimposing corner avoidance optimization adjustment, thereby achieving the technical effect of high-precision tool path planning, ensuring machining quality, controlling machining risks and improving machining efficiency. Attached Figure Description

[0008] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0009] Figure 1 This is a flowchart illustrating a high-precision toolpath planning method that considers material deformation compensation, as provided in an embodiment of this application.

[0010] Figure 2 This is a schematic diagram of a high-precision toolpath planning system that takes into account material deformation compensation, provided in an embodiment of this application.

[0011] Explanation of reference numerals in the attached diagram: Material deformation characteristics acquisition module 1, target tool acquisition module 2, initial tool path acquisition module 3, first result acquisition module 4, second result acquisition module 5. Detailed Implementation

[0012] This application provides a high-precision toolpath planning method and system that considers material deformation compensation, which solves the technical problem that traditional machining path planning methods fail to adapt to key influencing factors in actual machining, resulting in insufficient machining accuracy, uncontrollable risks, and poor efficiency.

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of this application, and not all of them. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.

[0014] It should be noted that the terms "first," "second," etc., in the specification and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of this application described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion; for example, a process, method, system, product, or server that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or modules not explicitly listed or inherent to such processes, methods, products, or devices.

[0015] Example 1, as Figure 1 As shown, a high-precision toolpath planning method considering material deformation compensation is disclosed, wherein the method includes:

[0016] According to the workpiece design scheme, the upper and lower composite inspections of the workpiece to be processed are carried out, the workpiece size difference mesh is established, and multi-directional material deformation finite element analysis is performed on the workpiece to be processed based on the workpiece size difference mesh to obtain the multi-directional material deformation characteristics.

[0017] Specifically, the process begins with simultaneous top-to-bottom detection of the workpiece to be processed to obtain top and bottom point cloud datasets. Then, a workpiece design model is constructed based on the workpiece design scheme. The obtained top and bottom point cloud datasets are then projected onto the workpiece design model to construct a design point cloud feature registration mesh. Finally, based on this design point cloud feature registration mesh, point-by-point design point cloud deviation detection is performed to generate a workpiece size difference mesh. This step will be explained in detail later.

[0018] Then, the generated workpiece size difference mesh is imported into commonly used finite element analysis software. Simultaneously, material parameters such as the elastic modulus, Poisson's ratio, and yield strength of the workpiece to be processed are entered into the software, completing the basic data configuration for the analysis model. The mesh integrity is then checked to ensure there are no mesh breaks or overlaps, providing a reliable model foundation for subsequent analysis. Based on the actual machining and clamping method of the workpiece, corresponding boundary constraints are set in the finite element analysis software to clearly define the degree of freedom restrictions in the fixed area of ​​the workpiece, avoiding unexpected displacements during the analysis process. Combining the cutting parameters of the tool on the workpiece in the machining process, the magnitude and direction of the cutting load are obtained through empirical formulas for calculating cutting forces. These loads are then applied to the corresponding machining areas of the workpiece size difference mesh according to the temporal distribution of the machining path, simulating the actual machining stress state.

[0019] Subsequently, multi-directional deformation analysis dimensions were set in the finite element analysis software, covering the three orthogonal directions (X-axis, Y-axis, and Z-axis) and oblique directions at different angles that may cause deformation during workpiece machining. An explicit or implicit solver suitable for structural deformation analysis was selected, and based on the fundamental equations of elasticity finite element analysis, the deformation process of the workpiece under cutting load was numerically iterated until convergence and stable analysis results were obtained. After the solution was completed, data such as displacement and strain values ​​of each mesh element in the preset multi-directional directions were extracted from the finite element analysis software. By organizing and classifying these data, the deformation differences in different regions and directions were analyzed, and key information such as the deformation patterns, maximum deformation, and deformation concentration areas of the workpiece along each direction during machining was summarized. Finally, a multi-directional material deformation characteristic that comprehensively reflects the deformation characteristics of the workpiece during machining was formed.

[0020] Construct a material property matrix for the workpiece to be processed, and select the appropriate tool from the tool library based on the workpiece design scheme to determine the target tool.

[0021] Optionally, first, the workpiece material data corresponding to the workpiece to be processed is loaded. Then, the loaded workpiece material data is sequentially cleaned and matrixed to generate the required workpiece material characteristic matrix. Next, the usage log sets corresponding to multiple tools in the tool library are retrieved. Then, based on the workpiece material characteristic matrix and the workpiece design scheme, the trigger frequency of each tool usage log set is statistically analyzed to obtain the workpiece material trigger value and workpiece design trigger value for each tool. Subsequently, the two types of trigger values ​​are weighted and fused according to a predetermined trigger weight to obtain the fitness of each tool. Finally, the tool library is iteratively optimized based on the fitness of each tool to determine the target tool. The above steps will be explained in detail in the following sections.

[0022] A machining process reinforcement chain is constructed, and the target tool path is planned according to the workpiece design scheme and the machining process reinforcement chain to obtain the initial tool path.

[0023] In one embodiment of this application, firstly, the corresponding processing technology chain is matched according to the workpiece design scheme. Then, based on the workpiece material property matrix, static risk assessment and dynamic risk assessment are performed on each processing node in the processing technology chain, and the node static risk sequence and node dynamic risk sequence are obtained in sequence. Finally, the processing technology chain is enhanced by multi-head attention by combining the two risk sequences to generate a processing technology enhancement chain. This step will be described in the following content.

[0024] Next, the core information of the workpiece design scheme and machining process reinforcement chain is analyzed. Key machining features of the metal workpiece to be machined are extracted from the workpiece design scheme, including contour boundaries, hole coordinates, groove dimensions, step height, and tolerance requirements. Simultaneously, the machining node sequence, optimized cutting parameter ranges for each node, cooling and lubrication requirements, and risk control points are obtained from the machining process reinforcement chain. These two types of information are categorized and integrated according to machining logic to form a basic dataset for path planning. Then, a machining coordinate system is established using the datum plane calibration method. Taking the datum plane or key positioning holes in the workpiece design model as references, the origin and X, Y, and Z axes of the coordinate system are set using a three-point plane determination method. This ensures precise alignment between the machining coordinate system and the coordinate system of the workpiece design model, providing a unified benchmark for toolpath coordinate calculation.

[0025] Next, based on the machining stages defined by the machining process enhancement chain, such as roughing, semi-finishing, and finishing, corresponding machining strategies are selected according to the machining objectives of each stage. For example, the roughing stage aims to efficiently remove excess metal and employs a layer-cutting strategy, cutting in layers according to a preset cutting depth, with each layer's cutting path offset along the workpiece contour. The semi-finishing stage aims to correct shape errors after roughing and employs a contour milling strategy to optimize the cutting path density and uniformly remove remaining material. The finishing stage aims to meet accuracy requirements and employs a climb milling strategy to reduce the impact of cutting forces on machining accuracy. The strategies for each stage are selected with reference to the standard machining strategy library built into the CAM software.

[0026] Subsequently, based on the basic parameters of the target tool, such as tool diameter, number of cutting edges, and cutting edge length, as well as the material properties of the metal workpiece and the parameter requirements of the machining process reinforcement chain, specific toolpath parameters are set. The cutting speed is determined according to the compatibility between the tool material and the metal material, referring to the recommended parameter range provided by the tool manufacturer and industry machining manuals. For example, when machining carbon steel with carbide tools, the cutting speed is set to 100-150 m / min. The feed rate is calculated as the product of the feed per tooth and the number of cutting edges, and the feed per tooth is selected as 0.1-0.3 mm / tooth according to the machining accuracy requirements. The depth of cut is set in combination with the tool rigidity and the workpiece allowance, taking 3-5 mm for roughing, 0.5-1 mm for semi-finishing, and 0.1-0.3 mm for finishing. At the same time, the parameters corresponding to high-risk nodes in the machining process reinforcement chain are adjusted to reduce the cutting speed or feed rate to control the risk.

[0027] Next, the integrated basic dataset, machining coordinate system, machining strategy and parameters are input into CAM software, such as UG and Mastercam. The software uses contour offset algorithm to generate roughing and semi-finishing paths, spiral interpolation or linear interpolation algorithm to generate hole machining paths, contour fitting algorithm to generate finishing paths, and automatically adds tool entry and exit trajectories. When entering the workpiece, a circular arc transition is used to avoid impacting the workpiece, and when exiting the workpiece, a straight line extension is used to leave the machining area, forming the preliminary tool path for each machining node.

[0028] Next, the collision interference detection function built into the CAM software is used to simulate and verify the preliminary tool path. The entire process of the target tool moving along the path is simulated to check whether there is a collision or interference between the tool and the workpiece and fixture. If a collision or interference is detected, the path is corrected by adjusting the tool posture, optimizing the coordinates of the path inflection points, or adding avoidance segments until all collision and interference risks are eliminated.

[0029] Finally, the modified toolpaths for each machining stage are integrated according to the node sequence of the machining process reinforcement chain, and a unified toolpath file is generated according to the format requirements of the machine tool control system. This file contains information such as the coordinate instructions, speed instructions, and feed instructions for tool movement, which is the initial toolpath that can be used for actual machining.

[0030] Based on the multi-directional material deformation characteristics, the initial toolpath is compensated for material deformation to obtain an adjustable toolpath space. The adjustable toolpath space is then optimized hierarchically using a machining risk assessment model to obtain the first result of toolpath optimization.

[0031] Specifically, firstly, based on the multi-directional material deformation characteristics results, the workpiece is divided into spatial coordinate zones. The maximum and average deformation values ​​in the X, Y, Z axes and tangential direction for each zone are extracted. These deformation data are then mapped one-to-one with the coordinate range of the corresponding zone, forming a zone deformation data lookup table. This ensures that the deformation information for each machining area is clearly available, providing a clear basis for compensation calculations. Next, the coordinate data of the initial toolpath is analyzed to obtain the three-dimensional coordinates of all continuous path points on the path. And the tangential angle. The coordinates of each path point are compared with the coordinate range of the partitioned deformation data in the partitioned deformation data lookup table to determine the workpiece partition to which each path point belongs, and then the deformation amount in each direction corresponding to that path point is extracted. , where ΔTn is the tangential deformation, and a unique correspondence is established between the initial toolpath point and the deformation.

[0032] Then, the reverse offset compensation method is used to perform specific compensation calculations. For each path point, the compensated path point parameters are calculated based on its corresponding deformation in each direction: Compensated X-coordinate = Initial X-coordinate - The compensated Y-coordinate = initial Y-coordinate - The compensated Z-coordinate = initial Z-coordinate - The compensated tangential angle = initial tangential angle - The process involves offsetting the deformation errors that may occur after machining the metal workpiece using reverse offsetting, ensuring that the final machined dimensions meet design requirements. Then, a partitioned smoothing method is used to optimize the compensation path. Linear interpolation is performed on the coordinates of the compensated path points within each partition to eliminate abrupt inflection points between path points, ensuring the continuity and stability of tool movement. For high-deformation partitions with deformation greater than 0.005mm, the path point density is appropriately increased to improve compensation accuracy. For medium-deformation partitions with deformation between 0.001mm and 0.005mm, the basic path point density is maintained consistent with the initial tool path, while the path point coordinates are slightly adjusted using a linear smoothing algorithm to eliminate minor coordinate fluctuations between adjacent path points, balancing machining stability and efficiency while meeting design tolerance requirements. For low-deformation partitions with deformation less than 0.001mm, the original path point density is maintained to balance machining accuracy and efficiency.

[0033] The rationality of the compensated toolpaths was then verified using a path simulation verification method. All compensated path points were connected according to the machining sequence, and the tool's movement along the path was simulated using CAM software to check whether the path exceeded the machine tool's travel range or interfered with the fixture or non-machined areas of the workpiece. If interference or travel exceeding limits were found, the compensation amount was readjusted for the corresponding path points until the path met the motion safety requirements. Finally, all compensated, smoothed, and verified toolpaths were integrated, and a path adjustment allowance was set at ±5% of the deformation in each direction, forming a set of alternative toolpaths with multiple different compensation allowances. This set constitutes the toolpath adjustment space.

[0034] Finally, a machining risk assessment model is used to optimize the toolpath space and establish a first optimal toolpath set. Machining quality risk and tool wear risk are then used as workpiece machining coupling risk factors, weighted and a workpiece machining coupling risk model is established. Subsequently, the coupling risk of the first optimal toolpath set is calculated and optimized based on this model. A second optimal toolpath set with coupling risk values ​​less than a set threshold is selected. Finally, the second optimal toolpath set is used to maximize machining efficiency, generating the first result of toolpath optimization. This step will be explained in detail later.

[0035] The first result of the toolpath optimization is applied to the adaptive angle avoidance compensation of the target tool to obtain the second result of the toolpath optimization.

[0036] Specifically, the first step involves analyzing the corner features of the toolpath optimization result to obtain multiple node corner features. Based on the target tool's basic tool data, tool yield prediction, chatter prediction, and overcut prediction are performed for each node's corner feature, sequentially obtaining the corner yield characteristics, chatter characteristics, and overcut characteristics of each node. Finally, combining the multiple node corner features with the above three types of characteristics, corner-related transition compensation optimization is performed on the first toolpath optimization result to generate the second toolpath optimization result. This step will be explained in detail later.

[0037] Furthermore, the method provided in this application embodiment includes:

[0038] The workpiece to be processed is subjected to synchronous detection at both the top and bottom to obtain the top and bottom point cloud datasets; a workpiece design model is constructed according to the workpiece design scheme; the top and bottom point cloud datasets are projected onto the workpiece design model to construct a design point cloud feature registration grid; point-by-point design point cloud deviation detection is performed according to the design point cloud feature registration grid to generate the workpiece size difference grid.

[0039] In this embodiment, the workpiece design model is a digital model built using tools such as CAD based on the workpiece design scheme. It accurately reflects the three-dimensional dimensions, structural shape, and characteristic parameters of the workpiece and serves as a reference benchmark for dimension comparison and path planning during the processing.

[0040] Specifically, when performing simultaneous top and bottom detection on the workpiece, two laser scanning sensors are symmetrically arranged above and below the workpiece. The two sensors start simultaneously with the same sampling frequency and preset scanning path to perform omnidirectional scanning of the upper and lower surfaces of the workpiece, capturing the three-dimensional coordinate information of the workpiece surface and forming discrete point sets. These are then organized to obtain the upper surface point cloud dataset and the lower surface point cloud dataset, which together form the upper and lower point cloud datasets.

[0041] Next, by receiving workpiece design drawings, technical specifications, and other documents from the design department, core information such as the workpiece's three-dimensional dimensions, structural shape, feature parameters, and processing requirements is obtained to form a workpiece design scheme. After obtaining the workpiece design scheme, using CAD modeling software, the workpiece's three-dimensional dimensions, structural shape, and feature parameters contained in the design scheme are imported into the software. According to the requirements of the workpiece design drawings, the workpiece's outline, holes, grooves, and other key structures are drawn sequentially to restore the workpiece's design form. Finally, a three-dimensional workpiece design model that is completely consistent with the workpiece design scheme is constructed, providing a benchmark for subsequent point cloud registration.

[0042] For the acquired upper and lower point cloud datasets, preprocessing is first performed using algorithms well-known to those skilled in the art, such as statistical filtering, Gaussian filtering, or direct-pass filtering. Statistical filtering removes isolated noise points that deviate from the average distance of neighboring points by more than a preset threshold. Gaussian filtering smooths the data with a weighted average to reduce the influence of random noise. Direct-pass filtering removes redundant points that exceed the reasonable size range of the workpiece according to a preset coordinate range, thereby ensuring the accuracy of the point cloud data.

[0043] Subsequently, the ICP iterative nearest point algorithm is used to first initialize the corresponding point pairs between the point cloud data and the workpiece design model. Then, the rigid transformation matrix that minimizes the sum of squared distances between corresponding point pairs is calculated. The transformation matrix is ​​iteratively updated until the number of iterations reaches a preset value or the distance error between point pairs is less than a threshold, thus achieving precise alignment of the preprocessed upper and lower point cloud datasets with the features of the 3D workpiece design model. After alignment, based on the workpiece's machining accuracy requirements and the structural complexity of the design model, a uniform mesh density is set according to industry-standard mesh generation, for example, setting the mesh side length to 0.01-0.1 mm. Regular mesh cells are constructed on the surface of the workpiece design model, with each mesh cell corresponding to a set of point cloud data, forming a design point cloud feature registration mesh, thus realizing the association between the point cloud data and the workpiece design model.

[0044] Finally, based on the point cloud feature registration mesh, the Euclidean distance calculation method is used to detect each point in the mesh one by one. For each point cloud data point in the mesh cell, the distance between its actual 3D coordinates and the theoretical coordinates of the corresponding workpiece design model mesh cell is calculated to obtain the dimensional deviation value of each point, and the deviation direction is recorded. The deviation data of all points are integrated according to their positions in the mesh, and the actual deviation value of each mesh cell is used to replace the original theoretical coordinate parameters, finally generating a workpiece dimensional difference mesh that can completely reflect the difference between the actual size and the design size of the workpiece to be processed.

[0045] By employing a series of steps including laser scanning to acquire point clouds, CAD software modeling, ICP registration and meshing, and point-by-point Euclidean distance deviation calculation, the difference between the actual size and the design size of the workpiece to be processed was accurately quantified, providing a reliable data foundation for subsequent multi-directional material deformation finite element analysis.

[0046] Furthermore, the method provided in this application embodiment includes:

[0047] Load the workpiece material data of the workpiece to be processed; clean and matrix the workpiece material data to generate the workpiece material characteristic matrix.

[0048] Optionally, the basic material data of the workpiece to be processed is first collected. The data sources include technical specifications provided by material suppliers, industry standard material databases, and measured data obtained from laboratory testing. The data covers key parameters related to processing deformation and cutting performance, such as elastic modulus, Poisson's ratio, yield strength, thermal conductivity, and hardness. Data import tools are used to import these scattered material data into the data processing software according to parameter type, thus completing the loading of workpiece material data.

[0049] Next, the workpiece material data after loading is cleaned, the reasonable value range of each parameter is selected, and abnormal data that exceeds the range is removed. For a small amount of missing data, linear interpolation is used to complete it. At the same time, redundant data with duplicate records is removed. By verifying the validity of each parameter one by one, the accuracy of the cleaned data and its compliance with the processing analysis requirements are ensured.

[0050] During subsequent matrix processing, the structure of the workpiece material property matrix is ​​first preset. The material parameter type is used as the row vector of the matrix, and the specific value of the corresponding parameter is used as the column vector of the matrix. The row vectors are arranged according to the priority order of parameters commonly used in elasticity and cutting engineering. All cleaned material parameters are sequentially filled into the preset structure. Through the matrix generation function of the data processing software, the structured material data is transformed into a two-dimensional numerical matrix. Finally, a workpiece material property matrix that can intuitively reflect the key characteristics of the workpiece material is generated, providing standardized data support for subsequent steps such as tool selection and machining risk assessment.

[0051] Furthermore, the method provided in this application embodiment includes:

[0052] The tool library retrieves multiple tool usage log sets corresponding to multiple tools; it performs trigger frequency statistics on each tool usage log set based on the workpiece material characteristic matrix to obtain the workpiece material trigger value for each tool; it performs trigger frequency statistics on each tool usage log set based on the workpiece design scheme to obtain the workpiece design trigger value for each tool; it performs weighted fusion of the workpiece material trigger value and the workpiece design trigger value for each tool based on a predetermined trigger weight to obtain the fitness of each tool; and iteratively optimizes the tool library based on the fitness of each tool to obtain the target tool.

[0053] In this embodiment of the application, the tool library is a digital database or physical storage and management system that stores parameter information, usage logs and other data of various metalworking tools for tool adaptation and selection.

[0054] Specifically, the process begins by retrieving a collection of usage logs for multiple tools stored in the tool library. Each tool's usage log contains detailed information such as the type of metal material it has processed in the past, the workpiece's design and structural features, processing parameters, and usage effects, ensuring that the retrieved log data fully covers relevant information related to the tool's application scenario.

[0055] Next, based on the workpiece material property matrix constructed in the previous steps, the core characteristic parameters such as hardness, elastic modulus, and yield strength of the workpiece to be processed are extracted. First, industry-standard allowable error thresholds are set for each core characteristic parameter. Then, the usage log set of each tool is analyzed one by one. The core characteristic parameters of the metal materials processed by the tool in the log are compared with the parameters corresponding to the workpiece to be processed. When the difference of all core characteristic parameters is within the respective preset allowable error threshold range, it is determined that a match is successful. Each successful match is counted once. Finally, the total number of matches is used as the workpiece material trigger value of the tool.

[0056] Then, based on the workpiece design scheme, the structural features of the metal workpiece to be processed are clarified, including key design information such as contour shape, hole distribution, groove size, and machining accuracy requirements. First, similarity judgment criteria are set for each key design feature. The contour shape is calculated for overlap using a geometric similarity algorithm and a similarity threshold is set. The hole distribution is calculated for deviation based on the number of holes and their relative position coordinates and a position deviation threshold is set. The groove size is compared with parameters such as depth and width and a dimensional error threshold is set. The machining accuracy requirements are judged based on whether the tolerance range is consistent. Similarly, the usage log set of each tool is checked one by one to determine whether the workpiece processed by the tool in the log meets the corresponding similarity judgment criteria for all key design features. When all design features meet the similarity requirements, it is judged as a similar design match. Each time it meets the requirements, a count is performed. The total number of matches is the workpiece design trigger value for that tool.

[0057] The adaptability of each tool is then calculated using a weighted summation formula. The specific formula is: Tool adaptability = workpiece material trigger value × tool-workpiece material trigger weight + workpiece design trigger value × tool-workpiece design trigger weight. This formula integrates the two dimensions of material adaptability and design adaptability through predetermined trigger weights to obtain a quantitative value that comprehensively reflects the tool adaptability. The predetermined trigger weights will be explained in detail later.

[0058] Finally, the fitness values ​​of all tools in the tool library are iteratively compared. The fitness of each tool is compared with the currently recorded maximum fitness value. The maximum fitness value and the corresponding tool information are continuously updated and recorded until the fitness comparison of all tools is completed. Finally, the tool corresponding to the maximum fitness value is determined as the target tool.

[0059] By retrieving logs from the database, obtaining trigger values ​​through traversal statistics, calculating fitness through weighted summation, and iteratively comparing and optimizing through traversal, a coherent process was achieved, which enabled precise matching of the material properties and design schemes of the cutting tool and the metal workpiece to be processed. This provided a highly adaptable tool foundation for subsequent high-precision toolpath planning.

[0060] Furthermore, the method provided in this application embodiment includes:

[0061] The predetermined trigger weights include tool and workpiece material trigger weights and tool and workpiece design trigger weights.

[0062] In one embodiment, the core factors affecting tool fit in metal processing are first identified, clarifying that the setting of tool-workpiece material trigger weights and tool-workpiece design trigger weights must be based on the degree of influence of material properties and design features on processing quality, efficiency, and tool life. Historical processing data from similar metal processing scenarios in the industry are collected, including tool failure rates and processing accuracy achievement rates under different material properties such as hardness and elastic modulus, as well as tool fit success rates and processing cost data corresponding to different design features such as complex contours and hole density, to establish a basic dataset for weight analysis.

[0063] Next, the weights were set using the Analytic Hierarchy Process (AHP). A hierarchical model was constructed, consisting of a target layer, a criterion layer, and a solution layer. The target layer represented "optimizing tool fit," the criterion layer represented "material fit influence" and "design fit influence," and the solution layer represented the corresponding two weight parameters. Based on collected historical data, 3-5 experienced metalworking technicians were invited to conduct pairwise comparisons of the relative importance of material fit influence and design fit influence in the criterion layer. A judgment matrix was generated using a 1-9 scale, where 1 indicates equal importance and 9 indicates extreme importance. For example, when material properties significantly influence tool life more than design features, a higher scale value was assigned to the material fit influence.

[0064] Next, the largest eigenvalue and its corresponding eigenvector of the judgment matrix are calculated through matrix operations. The eigenvector is then normalized to obtain the preliminary trigger weights for tool and workpiece materials and tool and workpiece design. A consistency check is then performed, calculating the consistency index CI and the consistency ratio CR. If CR is less than 0.1, the weight allocation is considered reasonable and no adjustment is needed; if CR is greater than or equal to 0.1, the judgment matrix is ​​reconstructed until the consistency requirements are met.

[0065] Finally, the initial weights are fine-tuned based on specific processing requirements. If the metal material in the processing scenario has high hardness and toughness, placing extremely high demands on the tool material and strength, the tool-workpiece material trigger weight can be appropriately increased. If the workpiece to be processed has a complex design structure with difficult-to-machine features such as dense holes and irregular grooves, requiring higher adaptability of the tool's cutting path, then the tool-workpiece design trigger weight can be appropriately increased. The sum of the two final determined weight values ​​is 1, forming a predetermined trigger weight that meets the requirements of the actual processing scenario.

[0066] Furthermore, the method provided in this application embodiment includes:

[0067] Based on the workpiece design scheme, a processing technology chain is matched; a static risk assessment is performed on each processing node in the processing technology chain based on the workpiece material property matrix to obtain a node static risk sequence; a dynamic risk assessment is performed on each processing node based on the workpiece material property matrix to obtain a node dynamic risk sequence; and multi-head attention enhancement is performed on the processing technology chain based on the node static risk sequence and the node dynamic risk sequence to generate the processing technology enhancement chain.

[0068] Specifically, the core information in the workpiece design scheme is first analyzed, including the structural complexity of the metal workpiece, key feature processing requirements, dimensional tolerance range, and overall processing difficulty. A metal processing technology database is then invoked. This database stores standard process flows, processing parameters, suitable materials, and workpiece characteristics for various metal processing tasks, providing a digital database for process matching and optimization. Next, a feature similarity matching algorithm is used to compare the key parameters in the workpiece design scheme with the parameters of various standard processing technology chains stored in the database. The process chain that best matches the design features of the workpiece is selected. Assuming this chain includes multiple processing nodes arranged in logical order, such as blanking, roughing, semi-finishing, finishing, hole machining, and surface treatment, the initial processing technology chain is formed.

[0069] Then, based on the constructed workpiece material property matrix, key parameters such as hardness, yield strength, thermal conductivity, and machinability of the metallic materials are extracted. For each processing node in the processing flow chain, static risk assessment indicators are determined, including the compatibility risk between the tool and the material, the inherent defect risk of the processing technology itself, and the processing difficulty risk caused by material properties. A risk level quantitative scoring method is adopted, referring to the industry's metal processing risk assessment standards, and a scoring range of 1-5 points is set for each indicator. Scores are assigned based on the compatibility between the material parameters and the processing node. The scores of each indicator are weighted and summed to obtain the static risk value of each processing node. All static risk values ​​are arranged sequentially according to the order of processing nodes to generate a node static risk sequence.

[0070] Subsequently, based on the workpiece material property matrix, the dynamic characteristics of metallic materials during processing are focused on, such as work hardening effect, thermal deformation trend, and stress accumulation law, to determine dynamic risk assessment indicators. These indicators include the risk of the impact of previous processing on subsequent processing, the risk of processing parameters adapting to changes in material state, and the risk accumulation caused by continuous processing. Using causal analysis combined with historical data statistics, historical processing data of similar materials at the same processing nodes are retrieved to analyze the probability and degree of risk occurrence at each node during continuous processing. The product of probability and degree of impact is taken as the dynamic risk value for that node. All dynamic risk values ​​are then arranged in the order of processing nodes to form a node dynamic risk sequence.

[0071] Subsequently, a simplified multi-head attention enhancement method based on weighted fusion is adopted. First, reasonable weights are assigned to the static and dynamic risk sequences of nodes according to the overall requirements of metal processing, with the sum of the weights being 1. If the processing scenario prioritizes process stability, the static risk weight is increased; if the continuity of continuous processing is prioritized, the dynamic risk weight is increased. The comprehensive risk value for each processing node is calculated as: Comprehensive Risk Value = Static Risk Value × Static Risk Weight + Dynamic Risk Value × Dynamic Risk Weight.

[0072] Finally, a comprehensive risk threshold is set: First, historical comprehensive risk values, corresponding processing quality compliance status, and tool wear levels are collected for each processing node in similar metal processing scenarios. The percentile method is used to statistically analyze the maximum comprehensive risk value in historical data where no processing quality defects have occurred and tool wear is within a reasonable range. This value is used as the initial threshold. Then, combined with the precision requirements of the current metal workpiece, material hardness and toughness characteristics, and the durability parameters of the target tool, the initial threshold is fine-tuned with reference to risk control standards in the metal processing industry. Finally, a comprehensive risk threshold suitable for the current processing scenario is determined. Next, for processing nodes with comprehensive risk values ​​exceeding this threshold, optimization measures are taken, including adjusting processing parameters to reduce cutting speed or feed rate, adding cooling or lubrication processes, inserting intermediate inspection nodes to verify processing effects, and adjusting the processing sequence of nodes to avoid risk accumulation. After optimizing and adjusting all processing nodes, they are logically integrated to form a reinforced processing flow chain.

[0073] By using a series of steps—feature matching to determine the process chain, quantitative scoring to obtain the risk sequence, and weighted fusion to enhance optimization—risk prediction and precise optimization of the machining process are achieved, providing a stable and reliable process foundation for subsequent path planning of the target tool.

[0074] Furthermore, the method provided in this application embodiment includes:

[0075] Based on the machining risk assessment model, the adjustment toolpath space is optimized for risk assessment to establish a first optimal toolpath set. The machining risk assessment index of the model is used as the workpiece machining coupling risk factor, which includes machining quality risk and tool wear risk. Weights are assigned to the workpiece machining coupling risk factor to establish a workpiece machining coupling risk model. Based on the workpiece machining coupling risk model, the first optimal toolpath set is optimized for workpiece machining coupling risk calculation to establish a second optimal toolpath set less than the workpiece machining coupling risk threshold. Based on the second optimal toolpath set, machining efficiency is maximized to generate the first result of the toolpath optimization.

[0076] Specifically, firstly, the m-th adjustment toolpath is extracted from the adjustment toolpath space and simulated to obtain the m-th machining simulation dataset, where m is a positive integer. The machining quality and tool wear correlation characteristic vector of the dataset is identified and input into the corresponding evaluation layer of the machining risk assessment model to obtain the corresponding risk value. If the two risk values ​​satisfy the machining risk constraint, the adjustment toolpath is added to the first optimal toolpath set. This step will be explained in detail later.

[0077] Next, the specific sources of workpiece machining coupling risk factors are clarified. Based on the established machining risk assessment model, the machining quality risk value Rq and tool wear risk value Rw corresponding to each tool path in the first optimal tool path set are extracted. Both of these values ​​are quantitative results in the 0-1 range and are the core basic data for constructing the workpiece machining coupling risk model.

[0078] Then, a weighted allocation method is used to set the weights for the two coupled risk factors. The weight allocation needs to be combined with the actual processing scenario requirements: if the processing scenario focuses on high precision, such as precision parts processing, then the processing quality risk weight 'a' is set to 0.6, and the tool wear risk weight 'b' is set to 0.4; if the processing scenario emphasizes the balance between precision and cost, such as mass production of parts, then 'a' is set to 0.5, and 'b' is set to 0.5; if the processing scenario has high requirements for tool wear control, such as processing expensive tools, then 'a' is set to 0.4, and 'b' is set to 0.6. The weights must satisfy the constraint that a + b = 1. Those skilled in the art can directly adjust the weight values ​​according to specific processing requirements.

[0079] Subsequently, a workpiece machining coupling risk model is established based on the linear weighting principle. The functional expression of the workpiece machining coupling risk model is: R = a × Rq + b × Rw, where R is the workpiece machining coupling risk value, a is the machining quality risk weight, b is the tool wear risk weight, Rq is the machining quality risk value, and Rw is the tool wear risk value. The coupling risk quantification result can be directly obtained through this formula, adapting to the practical requirements of industrial scenarios.

[0080] Next, the coupling risk is calculated for each toolpath in the first optimal toolpath set. The Rq and Rw values ​​of each path are substituted into the model to obtain the corresponding workpiece machining coupling risk value R. Then, a workpiece machining coupling risk threshold is set using the following method: Coupling risk value data of historical qualified toolpaths in similar metal processing scenarios are collected, with a sample size of no less than 50 groups. The 95th percentile value of the statistical data is used as the initial threshold. Then, fine-tuning is performed based on factors such as the precision level of the workpiece to be processed and tool cost. For example, the threshold can be lowered by 10% for high-precision workpieces and increased by 5% for low-cost tool processing. Finally, the workpiece machining coupling risk threshold is determined, with a typical range of 0.3-0.4. All toolpaths with a workpiece machining coupling risk value R less than this threshold are selected and integrated sequentially to form the second optimal toolpath set.

[0081] Finally, optimization to maximize machining efficiency was conducted, with machining time being identified as the core evaluation indicator; shorter machining time equates to higher efficiency. Machining time data for each toolpath in the second optimization toolpath set was extracted. This data could be directly obtained from the aforementioned simulation process, specifically the complete machining time recorded during the simulation. The machining times of all paths were sorted, and the toolpath with the shortest machining time was selected. This path represents the first result of toolpath optimization that balances machining risk control and machining efficiency.

[0082] By identifying risk factors, rationally allocating weights, constructing a simple coupled model, accurately screening thresholds, and ranking efficiency in a continuous process, hierarchical optimization of toolpaths was achieved. This ensured that the dual risks of machining quality and tool wear were controllable, while maximizing machining efficiency and providing an efficient and high-quality foundation path for subsequent corner avoidance compensation.

[0083] Furthermore, the method provided in this application embodiment includes:

[0084] The m-th adjustable tool path is extracted from the adjustable tool path space, and the workpiece to be processed is simulated according to the m-th adjustable tool path to obtain the m-th processing simulation dataset, where m is a positive integer; the processing quality correlation characteristics of the m-th processing simulation dataset are identified to obtain the m-th processing quality correlation characteristic vector, and the m-th processing quality correlation characteristic vector is input into the processing quality risk assessment layer embedded in the processing risk assessment model to obtain the m-th processing quality risk value; the tool wear correlation characteristics of the m-th processing simulation dataset are identified to obtain the m-th tool wear correlation characteristic vector, and the m-th tool wear correlation characteristic vector is input into the tool wear risk assessment layer embedded in the processing risk assessment model to generate the m-th tool wear risk value; it is determined whether the m-th processing quality risk value and the m-th tool wear risk value satisfy the processing risk constraint; if the m-th processing quality risk value and the m-th tool wear risk value satisfy the processing risk constraint, the m-th adjustable tool path is added to the first optimized tool path set.

[0085] In one embodiment, the m-th adjustable toolpath is first extracted sequentially from the adjustable toolpath space. Based on constraints such as the cutting parameter range and cooling / lubrication requirements in the machining process reinforcement chain, finite element simulation is performed on this path using CAM software. Key data are recorded during the simulation, including dimensional deviation data, surface roughness data, cutting force variation curves, cutting temperature data, and simulated tool edge wear. These data are integrated to form the m-th machining simulation dataset, where m is a positive integer. This simulation process is repeated for all adjustable toolpaths within this space.

[0086] Subsequently, the processing quality correlation characteristics of the m-th processing simulation dataset were identified. The core identification indicators were determined to be the maximum value of dimensional deviation, the standard deviation of dimensional deviation, the surface roughness Ra value, the shape tolerance value, and the position tolerance value. The specific values ​​of these five indicators were extracted from the dataset. The Min-Max normalization method was used to standardize the extracted indicator values, mapping the data to the 0-1 interval to eliminate dimensional differences. The processed 5-dimensional data vector is the m-th processing quality correlation characteristic vector.

[0087] Simultaneously, tool wear correlation characteristics are identified in the m-th machining simulation dataset. The core identification indicators are tool edge wear, peak cutting temperature, average cutting force, cutting time, and material removal rate. The specific values ​​of these five indicators are extracted from the dataset, and the Min-Max normalization method is used to standardize them into a 5-dimensional data vector in the 0-1 interval, which is the m-th tool wear correlation characteristic vector.

[0088] Next, a lightweight multilayer perceptron (MLP) model for machining risk assessment is constructed. The model comprises two parallel and independent assessment layers: a machining quality risk assessment layer and a tool wear risk assessment layer. Each assessment layer has a unified structure of "input layer-hidden layer-output layer": the input layer has 5 neurons, consistent with the dimension of the corresponding associated feature vector; the hidden layer has 10 neurons and uses the ReLU activation function to enhance the model's ability to fit nonlinear features; the output layer has 1 neuron and uses the Sigmoid activation function to map the output to the 0-1 range, facilitating risk quantification. The model as a whole adopts a dual-branch parallel architecture, with the two assessment layers sharing the training framework but having independent parameters, ensuring the specificity of their respective risk assessments.

[0089] The model training process is performed as follows: Historical machining data in the metal processing field is collected, containing at least 100 valid samples. Each sample must include tool path parameters, the corresponding machining simulation dataset, actual machining quality detection results, and tool wear detection results. Min-Max normalization is applied to the machining quality and tool wear related features in the samples. Simultaneously, the actual machining quality risk and tool wear risk are quantified into label values ​​in the 0-1 range according to industry standards. For example, a label value below 0.2 is set for acceptable machining quality, 0.4 for slight exceedance, 0.6 for moderate exceedance, and above 0.8 for severe exceedance; normal tool wear is set below 0.2, 0.5 for moderate wear, and above 0.8 for severe wear. The processed samples are divided into training and test sets in a 7:3 ratio. The training set is used for model parameter iteration, and the test set is used to verify model performance. Next, the training parameters are set. The Adam optimizer is selected, the learning rate is set to 0.001, the mean squared error (MSE) is used as the loss function, and the training epochs are set to 50 epochs. The model accuracy is verified once every 10 epochs using the test set. Training is stopped when the loss value on the test set no longer decreases for 3 consecutive epochs. The independent parameters of the two evaluation layers are saved, and the model construction is completed.

[0090] Then, the m-th machining quality correlation characteristic vector is input into the machining quality risk assessment layer embedded in the machining risk assessment model. It is then passed through the input layer to the hidden layer, where it is transformed by the ReLU activation function and mapped by the Sigmoid function of the output layer to obtain the m-th machining quality risk value Rq in the 0-1 interval. Similarly, the m-th tool wear correlation characteristic vector is input into the tool wear risk assessment layer, and the m-th tool wear risk value Rw in the 0-1 interval is obtained through the same calculation process.

[0091] Next, processing risk constraints are set: The processing quality risk constraint references the tolerance requirements of the workpiece design and industry processing quality standards, setting the threshold to 0.3, meaning a processing quality risk value ≤ 0.3 is considered to satisfy the constraint; the tool wear risk constraint is based on the durability parameters of the target tool and the company's processing cost control requirements, setting the threshold to 0.4, meaning a tool wear risk value ≤ 0.4 is considered to satisfy the constraint. The m-th processing quality risk value and the m-th tool wear risk value are jointly judged. If both values ​​satisfy the corresponding constraints, the m-th adjustment toolpath is added to the first optimized toolpath set.

[0092] Finally, repeat all the above steps to process all the adjustable toolpaths in the toolpath space in turn, where m traverses from 1 to the total number of adjustable toolpaths n, and completes the simulation machining, characteristic identification, risk value calculation and constraint judgment one by one until all paths are processed, and finally forms the first optimal toolpath set containing all adjustable toolpaths that meet the risk constraints.

[0093] By simulating machining data acquisition, feature standardization extraction, risk quantification and constraint screening using a lightweight MLP model, we achieved accurate risk assessment and efficient screening for adjusting toolpaths. We constructed a first set of optimal toolpaths that met the dual constraints of machining quality and tool wear, laying a reliable foundation for subsequent hierarchical optimization.

[0094] Furthermore, the method provided in this application embodiment includes:

[0095] The first result of toolpath optimization is subjected to corner feature analysis to obtain multiple node corner features; based on the basic tool data of the target tool, tool yield prediction is performed on each node corner feature to obtain the tool yield characteristics of each node corner; based on the basic tool data, chatter prediction is performed on each node corner feature to obtain the chatter characteristics of each node corner; based on the basic tool data, overcut prediction is performed on each node corner feature to obtain the overcut characteristics of each node corner; based on the multiple node corner features, and based on the tool yield characteristics, chatter characteristics, and overcut characteristics of each node corner, the first result of toolpath optimization is subjected to corner-related transition compensation optimization to generate the second result of toolpath optimization.

[0096] Optionally, the first result of toolpath optimization is analyzed for corner features. Based on the path geometry feature recognition algorithm built into the CAM software, the complete coordinate data and motion commands of the path are retrieved, nodes in the path whose direction changes are selected, and the three-dimensional coordinate information of the corner position of each node and the corner angle of the angle between two adjacent path segments are extracted. This information is then organized according to the path motion sequence to obtain multiple node corner features.

[0097] Next, the corner features of each node are predicted based on the basic tool data of the target tool. The specific prediction process is as follows:

[0098] Based on the technical principle of the correlation between cutting force and elastic deformation in metal processing, tool deflection prediction is conducted. Basic data of the target tool, including tool diameter, length, and material elastic modulus, are retrieved. Combined with the corner position, corner angle, and corresponding cutting parameters of each node's corner characteristics, the magnitude of the tool deflection is evaluated using a combination of cantilever beam deflection calculation and process system stiffness analysis. The average of the two methods is taken as the final tool deflection. Based on the degree of impact of the tool deflection on machining accuracy, the tool deflection characteristics are divided into three risk levels: slight tool deflection, moderate tool deflection, and severe tool deflection. This serves as the quantitative result of the tool deflection characteristics at each node's corner.

[0099] Chatter prediction is performed based on the technical logic of determining chatter based on the matching degree between cutting frequency and tool natural frequency. The natural frequency parameters of the target tool are extracted; these parameters can be directly obtained from the tool manufacturer's product manual. The actual cutting frequency at the corner is derived based on the cutting speed and tool number of cutting edges in the node corner characteristics. Using a frequency deviation rate calculation method, the matching degree between the actual cutting frequency and the tool natural frequency is compared, and chatter characteristics are classified into three levels: no chatter risk, slight chatter risk, and severe chatter risk, forming chatter characteristics for each node corner.

[0100] Overcut prediction is implemented based on geometric interference determination technology. Basic data of the target tool, including tool radius and tool tip radius, is retrieved. Simultaneously, the tool trajectory coordinates, workpiece design contour boundary coordinates at the corner, and corner type (concave or convex) are obtained from the corner features of each node. Using the geometric shortest distance calculation method, the actual distance from the tool center to the workpiece contour boundary is compared with the safe critical distance. Overcut characteristics are then classified into three levels: no overcut risk, slight overcut risk, and severe overcut risk, thus obtaining the overcut characteristics of each node corner.

[0101] Finally, based on the distribution logic of multiple node corner features, and combined with the three risk labels of tool deflection, chatter, and overcutting for each node, corner-related transition compensation optimization is carried out. For node corners with all three characteristics labeled as "slight" risk, the original path trajectory is maintained, and only linear smoothing is used to eliminate minor fluctuations. For node corners with any of the three characteristics labeled as "moderate" risk, the feed rate at the corner is adjusted or an arc transition is used to replace the original corner path to reduce the impact of sudden changes in cutting force. For node corners with any of the three characteristics labeled as "severe" risk, the tool center trajectory is recalculated. For convex corners, the tool center offset is appropriately increased, and for concave corners, a step-by-step machining method is adopted, i.e., rough machining is performed first to leave a margin, and then finish machining is performed to form the shape. At the same time, the characteristic correlation of adjacent node corners is considered to avoid new risks caused by compensation of a single node. The corner transition of the entire path is adapted and adjusted as a whole. Finally, all optimized path segments are integrated to generate the second result of tool path optimization.

[0102] Through a series of steps including corner feature analysis, three-factor risk prediction, and hierarchical correlation compensation optimization, the risks of tool deflection, chatter, and overcutting at toolpath corners are accurately avoided, improving the machining accuracy and motion stability of the toolpath and providing a reliable guarantee for the final high-precision machining.

[0103] In summary, the high-precision toolpath planning method considering material deformation compensation provided in this application has the following technical effects:

[0104] This application achieves the technical effect of high-precision toolpath planning, ensuring machining quality, controlling machining risks, and improving machining efficiency through workpiece design scheme, top-down composite inspection, establishing a dimensional differential mesh and analyzing multi-directional material deformation characteristics, constructing a material property matrix to select target tools, building a machining process reinforcement chain to plan the initial path, and through deformation compensation, hierarchical optimization and corner avoidance compensation.

[0105] Example 2, as Figure 2 As shown, based on the same inventive concept as in Embodiment 1, this application provides a high-precision toolpath planning system that considers material deformation compensation. The system includes:

[0106] Material deformation characteristic acquisition module 1 is used to perform upper and lower composite detection on the workpiece to be processed according to the workpiece design scheme, establish a workpiece size difference mesh, and perform multi-directional material deformation finite element analysis on the workpiece to be processed according to the workpiece size difference mesh to obtain multi-directional material deformation characteristics.

[0107] The target tool acquisition module 2 is used to construct the workpiece material property matrix of the workpiece to be processed, and to select the target tool from the tool library based on the workpiece design scheme.

[0108] The initial tool path acquisition module 3 is used to construct a machining process reinforcement chain and perform path planning for the target tool according to the workpiece design scheme and the machining process reinforcement chain to obtain the initial tool path.

[0109] First result acquisition module 4 is used to perform material deformation compensation on the initial tool path according to the multi-directional material deformation characteristics, obtain the adjustable tool path space, and perform hierarchical optimization on the adjustable tool path space through the machining risk assessment model to obtain the first result of tool path optimization.

[0110] The second result acquisition module 5 is used to perform adaptive angle avoidance compensation of the target tool on the first result of tool path optimization to obtain the second result of tool path optimization.

[0111] Furthermore, the material deformation characteristic acquisition module 1 is used to perform the following steps:

[0112] The workpiece to be processed is subjected to synchronous detection at both the top and bottom to obtain the top and bottom point cloud datasets; a workpiece design model is constructed according to the workpiece design scheme; the top and bottom point cloud datasets are projected onto the workpiece design model to construct a design point cloud feature registration grid; point-by-point design point cloud deviation detection is performed according to the design point cloud feature registration grid to generate the workpiece size difference grid.

[0113] Furthermore, the target tool acquisition module 2 is used to perform the following steps:

[0114] The tool library retrieves multiple tool usage log sets corresponding to multiple tools; it performs trigger frequency statistics on each tool usage log set based on the workpiece material characteristic matrix to obtain the workpiece material trigger value for each tool; it performs trigger frequency statistics on each tool usage log set based on the workpiece design scheme to obtain the workpiece design trigger value for each tool; it performs weighted fusion of the workpiece material trigger value and the workpiece design trigger value for each tool based on a predetermined trigger weight to obtain the fitness of each tool; and iteratively optimizes the tool library based on the fitness of each tool to obtain the target tool.

[0115] Furthermore, the initial toolpath acquisition module 3 is used to perform the following steps:

[0116] Based on the workpiece design scheme, a processing technology chain is matched; a static risk assessment is performed on each processing node in the processing technology chain based on the workpiece material property matrix to obtain a node static risk sequence; a dynamic risk assessment is performed on each processing node based on the workpiece material property matrix to obtain a node dynamic risk sequence; and multi-head attention enhancement is performed on the processing technology chain based on the node static risk sequence and the node dynamic risk sequence to generate the processing technology enhancement chain.

[0117] Furthermore, the first result acquisition module 4 is used to perform the following steps:

[0118] Based on the machining risk assessment model, the adjustment toolpath space is optimized for risk assessment to establish a first optimal toolpath set. The machining risk assessment index of the model is used as the workpiece machining coupling risk factor, which includes machining quality risk and tool wear risk. Weights are assigned to the workpiece machining coupling risk factor to establish a workpiece machining coupling risk model. Based on the workpiece machining coupling risk model, the first optimal toolpath set is optimized for workpiece machining coupling risk calculation to establish a second optimal toolpath set less than the workpiece machining coupling risk threshold. Based on the second optimal toolpath set, machining efficiency is maximized to generate the first result of the toolpath optimization.

[0119] Furthermore, the first result acquisition module 4 is used to perform the following steps:

[0120] The m-th adjustable tool path is extracted from the adjustable tool path space, and the workpiece to be processed is simulated according to the m-th adjustable tool path to obtain the m-th processing simulation dataset, where m is a positive integer; the processing quality correlation characteristics of the m-th processing simulation dataset are identified to obtain the m-th processing quality correlation characteristic vector, and the m-th processing quality correlation characteristic vector is input into the processing quality risk assessment layer embedded in the processing risk assessment model to obtain the m-th processing quality risk value; the tool wear correlation characteristics of the m-th processing simulation dataset are identified to obtain the m-th tool wear correlation characteristic vector, and the m-th tool wear correlation characteristic vector is input into the tool wear risk assessment layer embedded in the processing risk assessment model to generate the m-th tool wear risk value; it is determined whether the m-th processing quality risk value and the m-th tool wear risk value satisfy the processing risk constraint; if the m-th processing quality risk value and the m-th tool wear risk value satisfy the processing risk constraint, the m-th adjustable tool path is added to the first optimized tool path set.

[0121] Furthermore, the second result acquisition module 5 is used to perform the following steps:

[0122] The first result of toolpath optimization is subjected to corner feature analysis to obtain multiple node corner features; based on the basic tool data of the target tool, tool yield prediction is performed on each node corner feature to obtain the tool yield characteristics of each node corner; based on the basic tool data, chatter prediction is performed on each node corner feature to obtain the chatter characteristics of each node corner; based on the basic tool data, overcut prediction is performed on each node corner feature to obtain the overcut characteristics of each node corner; based on the multiple node corner features, and based on the tool yield characteristics, chatter characteristics, and overcut characteristics of each node corner, the first result of toolpath optimization is subjected to corner-related transition compensation optimization to generate the second result of toolpath optimization.

[0123] Furthermore, the target tool acquisition module 2 is used to perform the following steps:

[0124] Load the workpiece material data of the workpiece to be processed; clean and matrix the workpiece material data to generate the workpiece material characteristic matrix.

[0125] Furthermore, the target tool acquisition module 2 is used to perform the following steps:

[0126] The predetermined trigger weights include tool and workpiece material trigger weights and tool and workpiece design trigger weights.

[0127] The high-precision toolpath planning system considering material deformation compensation provided in the embodiments of the present invention can execute the high-precision toolpath planning method considering material deformation compensation provided in any embodiment of the present invention, and has the corresponding functional modules and beneficial effects of the execution method.

[0128] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0129] The specific embodiments described above do not constitute a limitation on the scope of protection of this application. Those skilled in the art should understand that various modifications, combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this application should be included within the scope of protection of this application. In some cases, the actions or steps described in this application can be performed in a different order than that shown in the embodiments and still achieve the desired results. Furthermore, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

Claims

1. A high-precision toolpath planning method considering material deformation compensation, characterized in that, The method includes: According to the workpiece design scheme, the upper and lower composite inspection of the workpiece to be processed is carried out, the workpiece size difference mesh is established, and multi-directional material deformation finite element analysis is performed on the workpiece to be processed based on the workpiece size difference mesh to obtain the multi-directional material deformation characteristics. Construct the workpiece material property matrix of the workpiece to be processed, and select the tool library based on the workpiece design scheme to determine the target tool; Construct a machining process reinforcement chain, and perform path planning for the target tool based on the workpiece design scheme and the machining process reinforcement chain to obtain an initial tool path; Based on the multi-directional material deformation characteristics, the initial toolpath is compensated for material deformation to obtain the adjustable toolpath space. The adjustable toolpath space is then optimized hierarchically using a machining risk assessment model to obtain the first result of toolpath optimization. The first result of the toolpath optimization is applied to the adaptive angle avoidance compensation of the target tool to obtain the second result of the toolpath optimization.

2. The high-precision toolpath planning method considering material deformation compensation as described in claim 1, characterized in that, Based on the workpiece design scheme, perform upper and lower composite inspections on the workpiece to be processed, and establish a workpiece dimensional difference mesh, including: The workpiece to be processed is simultaneously detected from top to bottom to obtain a top and bottom point cloud dataset; Based on the workpiece design scheme, construct the workpiece design model; The upper and lower point cloud datasets are projected onto the workpiece design model to construct a design point cloud feature registration mesh. Based on the design point cloud feature registration mesh, point-by-point design point cloud deviation detection is performed to generate the workpiece size difference mesh.

3. The high-precision toolpath planning method considering material deformation compensation as described in claim 1, characterized in that, Based on the workpiece design scheme, the tool magazine is selected for adaptability, and the target tool is determined, including: Retrieve multiple tool usage log sets corresponding to multiple tools in the tool library; Based on the workpiece material property matrix, the trigger frequency of each tool usage log set is statistically analyzed to obtain the workpiece material trigger value for each tool. Based on the workpiece design scheme, the trigger frequency of each tool usage log set is statistically analyzed to obtain the workpiece design trigger value for each tool. The material trigger values ​​and design trigger values ​​of each tool and workpiece are weighted and fused according to a predetermined trigger weight to obtain the adaptability of each tool. The tool library is iteratively optimized based on the fitness of each tool to obtain the target tool.

4. The high-precision toolpath planning method considering material deformation compensation as described in claim 1, characterized in that, Constructing a reinforced processing chain includes: Based on the workpiece design scheme, match the processing technology flow chain; Based on the workpiece material property matrix, a static risk assessment is performed on each processing node in the processing technology chain to obtain the node static risk sequence. Based on the workpiece material property matrix, a dynamic risk assessment is performed on each processing node to obtain a node dynamic risk sequence. Multi-head attention enhancement is performed on the processing flow chain based on the node static risk sequence and the node dynamic risk sequence to generate the processing flow enhancement chain.

5. A high-precision toolpath planning method considering material deformation compensation as described in claim 1, characterized in that, The toolpath space is hierarchically optimized using a machining risk assessment model to obtain the first result of toolpath optimization, including: Based on the machining risk assessment model, the adjustment toolpath space is optimized by risk assessment, and a first set of optimized toolpaths is established. The processing risk assessment index of the processing risk assessment model is used as the workpiece processing coupling risk factor, and the processing risk assessment index includes processing quality risk and tool wear risk; Weights are assigned to the workpiece processing coupling risk factors to establish a workpiece processing coupling risk model; Based on the workpiece machining coupling risk model, the first optimal tool path set is optimized by calculating the workpiece machining coupling risk, and a second optimal tool path set less than the workpiece machining coupling risk threshold is established. Based on the second set of optimized toolpaths, the machining efficiency is maximized to generate the first result of the toolpath optimization.

6. The high-precision toolpath planning method considering material deformation compensation as described in claim 5, characterized in that, Based on the machining risk assessment model, the adjustment toolpath space is optimized through risk assessment to establish a first set of optimized toolpaths, including: The m-th adjustable tool path is extracted based on the adjustable tool path space, and the workpiece to be processed is simulated based on the m-th adjustable tool path to obtain the m-th processing simulation dataset, where m is a positive integer; The processing quality correlation characteristics of the m-th processing simulation dataset are identified to obtain the m-th processing quality correlation characteristic vector. The m-th processing quality correlation characteristic vector is then input into the processing quality risk assessment layer embedded in the processing risk assessment model to obtain the m-th processing quality risk value. The tool wear correlation characteristics of the m-th machining simulation dataset are identified to obtain the m-th tool wear correlation characteristic vector. The m-th tool wear correlation characteristic vector is then input into the tool wear risk assessment layer embedded in the machining risk assessment model to generate the m-th tool wear risk value. Determine whether the m-th machining quality risk value and the m-th tool wear risk value satisfy the machining risk constraints; If the m-th machining quality risk value and the m-th tool wear risk value satisfy the machining risk constraint, the m-th adjustment toolpath is added to the first optimized toolpath set.

7. The high-precision toolpath planning method considering material deformation compensation as described in claim 1, characterized in that, The first result of toolpath optimization is used to perform adaptive angle avoidance compensation for the target tool to obtain the second result of toolpath optimization, including: The first result of the toolpath optimization is subjected to corner feature analysis to obtain the corner features of multiple nodes; Based on the basic tool data of the target tool, the tool deflection prediction is performed on the corner features of each node to obtain the tool deflection characteristics of each node corner. Based on the tool's basic data, chatter prediction is performed on the corner features of each node to obtain the chatter characteristics of each node's corner. Based on the tool's basic data, overcut prediction is performed on the corner features of each node to obtain the overcut characteristics of each node's corner. Based on the multiple node corner features, the tool path optimization first result is optimized by corner-related transition compensation according to the corner deflection characteristics, corner chatter characteristics, and corner overcut characteristics of each node, thereby generating the tool path optimization second result.

8. A high-precision toolpath planning method considering material deformation compensation as described in claim 1, characterized in that, Constructing the workpiece material property matrix of the workpiece to be processed includes: Load the workpiece material data of the workpiece to be processed; The workpiece material data is cleaned and matrixed to generate the workpiece material characteristic matrix.

9. A high-precision toolpath planning method considering material deformation compensation as described in claim 3, characterized in that, The predetermined trigger weights include tool and workpiece material trigger weights and tool and workpiece design trigger weights.

10. A high-precision toolpath planning system considering material deformation compensation, characterized in that, The system is used to implement the high-precision toolpath planning method considering material deformation compensation according to any one of claims 1-9, the system comprising: The material deformation characteristics acquisition module is used to perform upper and lower composite detection on the workpiece to be processed according to the workpiece design scheme, establish a workpiece size difference mesh, and perform multi-directional material deformation finite element analysis on the workpiece to be processed based on the workpiece size difference mesh to obtain multi-directional material deformation characteristics. The target tool acquisition module is used to construct the workpiece material property matrix of the workpiece to be processed, and to select the target tool from the tool library based on the workpiece design scheme. The initial toolpath acquisition module is used to construct a machining process reinforcement chain and perform path planning for the target tool based on the workpiece design scheme and the machining process reinforcement chain to obtain the initial toolpath. The first result acquisition module is used to perform material deformation compensation on the initial tool path according to the multi-directional material deformation characteristics, obtain the adjustable tool path space, and perform hierarchical optimization of the adjustable tool path space through the machining risk assessment model to obtain the first result of tool path optimization. The second result acquisition module is used to perform adaptive angle avoidance compensation of the target tool on the first result of tool path optimization to obtain the second result of tool path optimization.

Images