Precision machining and 3d printing hybrid manufacturing execution system

Abstract

The application provides a precision machining and 3D printing hybrid manufacturing execution system, and relates to the technical field of data processing.The system is used for: dividing a three-dimensional model into a configuration area and a finishing area, generating a layer-by-layer closed contour path, obtaining configuration path data, extracting a boundary segment close to the finishing area, and calculating a spatial projection vector of the boundary segment on a finishing target surface to obtain preliminary finishing path data, calculating a boundary twist value, superimposing an interference offset in a historical sample path to the preliminary finishing path, calculating an interference amplitude index, performing coordinate correction on the preliminary finishing path, adjusting the position and direction of each preliminary finishing path point, obtaining fitting path data, connecting and uniformly processing the fitting path data and the configuration path data according to a manufacturing sequence to obtain hybrid execution path instructions.The application integrates precision machining and 3D printing into one, and avoids path overlap, air cutting or unnecessary machining.

Description

Technical Field

[0001] This invention relates to the field of data processing technology, and in particular to a hybrid manufacturing execution system combining precision machining and 3D printing. Background Technology

[0002] In modern manufacturing processes, the combination of 3D printing and CNC precision machining is widely used in the customized production of highly complex parts. Typically, 3D printing is used for rough machining of the shape, and then CNC is used for fine machining of dimensional accuracy and surface quality. The manufacturing process is executed by two independent platforms. After 3D printing is completed, the model needs to be exported manually and the CNC path needs to be manually arranged to complete the switch from additive to subtractive manufacturing.

[0003] However, the lack of a joint optimization mechanism based on the CNC model in the machining path may lead to redundancy in the CNC path and mismatch with the actual remaining material distribution after 3D printing. For example, when non-standard boundaries are formed due to material shrinkage or stacking deviation during 3D printing, the CNC path still runs according to the standard model, which may not be able to adaptively adjust the tool path, resulting in problems such as redundant cutting, overcutting, or residue, and unnecessary waste of machining allowance. Summary of the Invention

[0004] The purpose of this invention is to provide a hybrid manufacturing execution system for precision machining and 3D printing, which aims to solve the problems mentioned in the background art.

[0005] To solve the above-mentioned technical problems, the technical solution of the present invention is as follows:

[0006] A hybrid manufacturing execution system combining precision machining and 3D printing, the system comprising:

[0007] The partitioning module is used to identify regions based on the 3D model of the target component, dividing the 3D model into configuration areas and trimming areas to obtain structural distribution data;

[0008] The edge-forming module is used to perform gradient fitting on continuous contour points within the configuration area based on structural distribution data, generate configuration boundaries, and generate layer-by-layer closed contour paths along the configuration boundaries to obtain configuration path data.

[0009] The mapping module is used to extract boundary segments close to the trimming area based on the configuration boundary data, and calculate their spatial projection vectors on the trimming target surface to obtain preliminary trimming path data.

[0010] The torsion module is used to extract the arc and length of the preliminary trimming path based on the preliminary trimming path data, and to calculate the boundary torsion value.

[0011] The perturbation module is used to superimpose the interference offset in the historical sample path onto the preliminary trimming path based on the boundary torsion value, and extract the amplitude peak and change gradient to calculate the interference amplitude index.

[0012] The correction module is used to correct the coordinates of the preliminary correction path according to the interference amplitude index, adjust the position and direction of each preliminary correction path point, and obtain the fitted path data.

[0013] The splicing module is used to connect and unify the paths according to the manufacturing sequence based on the fitted path data and the configuration path data to obtain mixed execution path instructions.

[0014] Furthermore, the partitioning module includes:

[0015] The node extraction unit is used to extract nodes from the 3D model and perform geometric distribution sampling to obtain discrete node data.

[0016] The curvature calculation unit is used to measure the local surface curvature of the three-dimensional model based on discrete node data to obtain curvature distribution data.

[0017] The segmentation unit is used to perform preliminary region classification of the 3D model based on curvature distribution data, forming the initial configuration segment and the initial trimming segment;

[0018] Boundary fusion units are used to reorganize the contour lines between the initial configuration segment and the initial trimming segment, adjust the node distribution of discontinuous boundaries, form configuration and trimming regions, and obtain structural distribution data.

[0019] Furthermore, the edge-forming module includes:

[0020] The contour extraction unit is used to extract contour points of the boundary of the configuration area based on the structural distribution data, and obtain contour point sequence data.

[0021] The gradient analysis unit is used to perform spatial direction change rate analysis on the contour point sequence data, calculate the tangential direction difference between adjacent contour points, and obtain contour gradient data.

[0022] The fitting building unit is used to fit the boundary lines of the contour points based on the contour gradient data to generate the configuration boundary.

[0023] The path generation unit is used to perform equal-layer scanning along the configuration boundary, extract the closed contour lines on each section, generate layer-by-layer processing paths, and obtain configuration path data.

[0024] Furthermore, the mapping module includes:

[0025] The boundary filtering unit is used to extract the boundary segments adjacent to the trimming area based on the spatial distance between the boundary segments and the trimming area, and obtain the set of adjacent boundary segments.

[0026] The projection vector unit is used to construct a normal projection vector for each boundary point in the adjacent boundary segment set, and to filter the direction based on the relationship between the projection direction and the position of the trimmed area to obtain the target projection vector set.

[0027] The spatial mapping unit is used to project the target projection vector set onto the trimmed target surface, establish the spatial correspondence between boundary points and projection points, and form a set of mapping line segments.

[0028] The path construction unit is used to sequentially connect the mapped line segment set to generate a continuous path line structure and obtain preliminary path data.

[0029] Furthermore, the torsion module includes:

[0030] The path segment radian unit is used to extract the included angle between any two adjacent preliminary trimming paths in the preliminary trimming path data, convert it into radian values, remove the units of measurement, and obtain the path radian data.

[0031] The curvature measurement construction unit is used to extract the length of each preliminary trimming path segment and, in combination with the path arc data, calculate the curvature radius of each preliminary trimming path segment to obtain curvature radius data.

[0032] The rate of change analysis unit is used to calculate the rate of change of arc between adjacent preliminary trimming paths based on the path arc data;

[0033] The boundary torsion value unit is used to calculate the boundary torsion value based on the path radii data, radius of curvature data, and rate of change of radii.

[0034] Furthermore, the boundary torsion value unit includes:

[0035] The boundary torsion value calculation unit is used to extract the direction angle value of each preliminary trimming path segment and fuse it with the path length and the radius of curvature of the corresponding segment to capture high-risk bending paths and obtain a bending density term; fuse the rate of change of radians with the radius of curvature of the corresponding segment to measure the instability in the continuous change process of the path and obtain a direction change rate term; fuse the rate of change of radians with the radius of curvature of the corresponding segment to suppress the drastic amplification caused by small curvature and obtain a curvature suppression term; and obtain an adjustment factor term based on the length ratio of the corresponding segments.

[0036] By fusing the suppression curvature term and the regulation factor term, the control discontinuity caused by abrupt changes between local path segments is identified, resulting in the inter-segment abrupt change modulation term. The bending density term, the normalization term for the rate of change of direction, and the inter-segment abrupt change modulation term are then fused to obtain the boundary torsion value.

[0037] Furthermore, the disturbance module includes:

[0038] The path matching unit is used to search for path segments with similar structures in the historical sample path database based on the structure and boundary torsion value of the initially modified path, and generate matching index pairs.

[0039] The interferometric overlay unit is used to extract the interferometric offset from historical samples based on the matching index and to perform vector overlay on the preliminary trimming path to obtain the perturbation fusion path.

[0040] The perturbation feature extraction unit is used to identify concentrated perturbation regions on the perturbation fusion path, extract the offset peak value and gradient change value of the path node, and obtain perturbation feature data.

[0041] The interference amplitude unit is used to construct an interference response function based on disturbance characteristic data, reflecting the overall severity of interference along the path and obtaining the interference amplitude index.

[0042] Furthermore, the interference amplitude unit includes:

[0043] The interference amplitude calculation unit is used to calculate the mean, standard deviation and gradient change of the path node offset, and then calculate the mean and standard deviation of the gradient change.

[0044] Based on the peak, mean, and standard deviation of the offset of the path nodes, and combined with their normalized relative positions in the path, the influence of a single outlier on the whole is suppressed, resulting in the amplitude peak normalization amplification term; based on the maximum and mean of the gradient change values ​​of the path nodes, the gradient strength of the path nodes in the overall offset is measured, resulting in the gradient strength comprehensive term; based on the standard deviation and mean of the gradient change values ​​of the path nodes, their coefficient of variation is calculated, and based on this, the relative dispersion of the gradient distribution is measured, resulting in the unbalanced modulation term.

[0045] The interference amplitude index is obtained by fusing the amplitude peak normalization amplification term, the gradient intensity comprehensive term, and the unbalanced modulation term.

[0046] Furthermore, the correction module includes:

[0047] The weight generation unit is used to allocate the magnitude according to the interference magnitude index, and to proportionally divide the interference magnitude index into a weight sequence corresponding to each path node to obtain weight data.

[0048] The offset calculation unit is used to calculate the coordinate increment based on the weight data and the distance between adjacent path nodes, calculate the offset increment of each path node on the three axes, and obtain the coordinate increment sequence.

[0049] The orientation calculation unit is used to adjust the orientation based on the coordinate increment sequence and the orientation sequence of the preliminary trimming path, generate orientation coefficients corresponding to each path node, and obtain the orientation coefficient sequence.

[0050] The point update unit is used to update the node position and direction vector according to the coordinate increment sequence and the direction coefficient sequence, so as to obtain the updated node set;

[0051] The constraint projection unit is used to perform boundary constraint projection based on the structural distribution data, and to revise the projection of out-of-bounds nodes in the updated node set to obtain a boundary compliant node set.

[0052] The path fitting unit is used to perform continuous curve fitting and inter-segment connection to generate a path segment sequence based on the boundary compliance node set, thereby obtaining the fitted path data.

[0053] Furthermore, the splicing module includes:

[0054] The sequential parsing unit is used to sequentially parse the fitted path data and configuration path data based on the timing information of the manufacturing process to obtain the path execution sequence data.

[0055] The endpoint matching unit is used to extract the start and end nodes of the fitted path data and the configuration path data based on the path execution order data, perform endpoint distance calculation and shortest connection matching, and obtain endpoint matching pairs.

[0056] The connection generation unit is used to splice path segments according to endpoint matching pairs, and insert transition segments at the splicing points to ensure curvature continuity, thereby obtaining mixed path data;

[0057] The format unification unit is used to convert mixed path data into a unified path data structure format, perform node index rearrangement and segment attribute labeling to obtain structured mixed path data, and translate it into a control instruction set that can be recognized by the manufacturing equipment to obtain mixed execution path instructions.

[0058] The above-described solution of the present invention has at least the following beneficial effects:

[0059] This invention identifies and divides the 3D model of a target component into a configuration area and a trimming area, transforming the originally continuous and complex 3D data structure into two substructures with manufacturing semantics. This achieves structural decomposition of the model data. Compared to the traditional manufacturing process where 3D model data enters the processing path generation stage as a whole, resulting in a lack of boundary understanding ability in the CNC system, this solution makes the subsequent path calculation logic more targeted through area division. It achieves hierarchical management of data in the spatial domain, effectively reduces data redundancy in the overall path planning, and improves the clarity and logical relevance in the data transmission link. This is a key step in the transition of manufacturing systems from geometry-driven to data structure-driven.

[0060] This invention extracts boundary segments near the trimming area from the configuration boundary and projects them onto the trimming target surface to form preliminary trimming path data. It establishes a data transformation relationship in the spatial domain, mapping from the configuration boundary coordinate system to the trimming path coordinate system. Traditional methods rely on standard geometric models for CNC trimming path extraction, failing to perceive the actual boundary errors after printing. This system extracts the actual configuration boundary and maps the trimming path using vector projection, achieving perception-driven adjustment of the path data. This ensures the preliminary trimming path is generated based on the actual configuration boundary, possessing geometric consistency with the printed shape, and solving the problem of path deviation caused by idealized modeling.

[0061] This invention calculates the curvature and length of the initial path adjustment to form a boundary torsion value. This index measures the nonlinearity of the path structure, constructing a numerical characteristic of path structural complexity that can quantitatively reflect the path's changing trend. By modeling the curvature characteristics of the path and introducing an evaluation mechanism, the boundary torsion value can accurately capture potential curvature abrupt changes or continuity failure points in the path segment, providing a reliable reference index for subsequent path disturbances and corrections. It can be used for path visualization evaluation, providing real-time feedback on path status in digital twin or path simulation systems, achieving data closed-loop control, and serving as typical data support for the intelligent transition of manufacturing path systems.

[0062] This invention extracts continuous contour points and performs gradient fitting to construct configuration boundaries, then generates layer-by-layer closed contour paths, ultimately forming configuration path data. It establishes a gradient progression relationship from point set data to path data, namely, extraction of boundary points - calculation of direction difference - path closure organization. This is significantly better than the traditional method of generating printing paths on a per-section-layer basis. The latter often leads to problems such as path breakage and layer non-closure due to irregular boundaries or disordered point order. The closed path ensures that each layer of configuration data has a complete trajectory description during the printing process, avoiding printing anomalies due to missing data.

[0063] This invention connects the configuration path and the fitting and trimming path through the manufacturing sequence and outputs hybrid execution path instructions, which manifests as the standardization and fusion of path data structures. In traditional hybrid manufacturing processes, 3D printing and CNC paths exist independently, and path files are often in different formats, requiring manual conversion and interfacing, which can easily lead to data misalignment. This system, through steps such as endpoint matching, transition segment generation, and format unification, realizes the translation of path data from non-uniform formats to a unified instruction set, ensuring that the manufacturing equipment can seamlessly switch between different process paths during execution and eliminating problems such as processing stagnation and path truncation caused by interface inconsistencies. Attached Figure Description

[0064] Figure 1This is a flowchart of a precision machining and 3D printing hybrid manufacturing execution system provided in an embodiment of the present invention. Detailed Implementation

[0065] Exemplary embodiments of the present disclosure will now be described in more detail with reference to the accompanying drawings. While exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

[0066] like Figure 1 As shown, embodiments of the present invention propose a hybrid manufacturing execution system for precision machining and 3D printing, the system comprising:

[0067] The partitioning module is used to identify regions based on the 3D model of the target component, dividing the 3D model into configuration regions and trimming regions to obtain structural distribution data;

[0068] The edge-forming module is used to perform gradient fitting on continuous contour points within the configuration area based on structural distribution data, generate configuration boundaries, and generate layer-by-layer closed contour paths along the configuration boundaries to obtain configuration path data.

[0069] The mapping module is used to extract boundary segments close to the trimming area based on the configuration boundary data, and calculate their spatial projection vectors on the trimming target surface to obtain preliminary trimming path data.

[0070] The torsion module is used to extract the arc and length of the preliminary trimming path based on the preliminary trimming path data, and to calculate the boundary torsion value.

[0071] The perturbation module is used to superimpose the interference offset in the historical sample path onto the preliminary trimming path based on the boundary torsion value, and extract the amplitude peak and change gradient to calculate the interference amplitude index.

[0072] The correction module is used to correct the coordinates of the preliminary correction path according to the interference amplitude index, adjust the position and direction of each preliminary correction path point, and obtain the fitted path data.

[0073] The splicing module is used to connect and unify the paths according to the manufacturing sequence based on the fitted path data and the configuration path data to obtain mixed execution path instructions.

[0074] In this embodiment of the invention, the partitioning module is used to identify regions based on the three-dimensional model of the target component, dividing the three-dimensional model into a configuration area and a trimming area to obtain structural distribution data. This achieves pre-division of the manufacturing area at the data level, effectively distinguishing the application boundaries of 3D printing and CNC trimming, and avoiding path overlap or omission. The edge-forming module is used to perform gradient fitting on continuous contour points within the configuration area based on the structural distribution data, generating configuration boundaries, and generating layer-by-layer closed contour paths along the configuration boundaries to obtain configuration path data, avoiding printing discontinuities due to discontinuous point sets. The mapping module is used to extract boundary segments near the trimming area based on the configuration boundary data, and calculate their spatial projection vectors on the trimming target surface to obtain preliminary trimming path data. This achieves geometric coordination and transformation of path data, enhancing the fit between the trimming path and the actual printing result.

[0075] The torsion module extracts the curvature and length of the preliminary trimming path from the preliminary trimming path data, calculates the boundary torsion value, helps identify high-risk path segments in advance, and provides input parameters for path disturbance adjustment. The disturbance module superimposes the interference offset from historical sample paths onto the preliminary trimming path based on the boundary torsion value, extracts the amplitude peak and change gradient, calculates the interference amplitude index, enabling the system to have predictive correction capabilities and improve the intelligent adaptation level of the path. The correction module corrects the coordinates of the preliminary trimming path based on the interference amplitude index, adjusts the position and direction of each preliminary trimming path point, obtains fitted path data, integrates disturbance information and maintains boundary legality, and improves the manufacturing executability and equipment adaptability of the path. The splicing module connects and unifies the paths according to the manufacturing sequence based on the fitted path data and the configuration path data to obtain mixed execution path instructions, realizes seamless connection of the manufacturing path from configuration to trimming, and improves the logical integrity of the path execution process.

[0076] In a preferred embodiment of the present invention, the partitioning module includes:

[0077] The node extraction unit is used to extract nodes from the 3D model and perform geometric distribution sampling to obtain discrete node data.

[0078] The curvature calculation unit is used to measure the local surface curvature of the three-dimensional model based on discrete node data to obtain curvature distribution data.

[0079] The segmentation unit is used to perform preliminary region classification of the 3D model based on curvature distribution data, forming the initial configuration segment and the initial trimming segment;

[0080] Boundary fusion units are used to reorganize the contour lines between the initial configuration segment and the initial trimming segment, adjust the node distribution of discontinuous boundaries, form configuration and trimming regions, and obtain structural distribution data.

[0081] In this embodiment of the invention, a node extraction unit is used to extract nodes from the 3D model and perform geometric distribution sampling to obtain discrete node data, realizing the transformation of the model from continuous representation to discrete data format, providing a foundation for subsequent data analysis and geometric operations; a curvature calculation unit is used to measure the local surface curvature of the 3D model based on the discrete node data to obtain curvature distribution data, providing a precise quantitative basis for subsequent region division; a segment division unit is used to perform preliminary region classification of the 3D model based on the curvature distribution data to form initial configuration segments and initial trimming segments, realizing semantic segmentation of the 3D model, and providing independent but complementary data support for configuration paths and trimming paths; a boundary fusion unit is used to reorganize the contour lines between the initial configuration segments and the initial trimming segments, adjust the node distribution of discontinuous boundaries, form configuration areas and trimming areas, and obtain structural distribution data, solving problems such as path breakage, splicing irregularities, and tool jump caused by discontinuous boundaries between configuration and trimming areas, laying the foundation for subsequent path concatenation in the splicing module.

[0082] The curvature calculation unit is used to measure the local surface curvature of the 3D model based on discrete node data, and obtain curvature distribution data, specifically including:

[0083] The system first receives a discrete node dataset output by the node extraction unit. Each node contains spatial coordinate information and its adjacent connectivity information. Based on this node set, a local neighborhood search structure is constructed, such as a local node set established based on the radius adjacency algorithm, to ensure that each node under test has support points in its geometric neighborhood. Then, the system performs a local surface fitting operation in the local neighborhood of each node using the least squares method. The fitting function can be in the form of a quadratic surface, constructing a local function of the form... The algebraic expression of .

[0084] By calculating the first and second partial derivatives of the locally fitted surface, the normal vector and principal curvature directions are obtained, respectively. This allows for the further calculation of the Gaussian curvature and mean curvature of each node. The Gaussian curvature is obtained by multiplying the principal curvatures, while the mean curvature is obtained by the arithmetic mean of the principal curvatures. To improve the stability of the curvature data, extreme points or isolated peaks within the node's neighborhood are smoothed. Finally, the curvature value corresponding to each node is bound to the node's spatial coordinates, forming curvature distribution data for subsequent partitioning and classification.

[0085] The boundary fusion unit is used to reorganize the contour lines between the initial configuration segment and the initial trimming segment, adjust the node distribution of discontinuous boundaries, form the configuration region and the trimming region, and obtain structural distribution data, specifically including:

[0086] The system first receives the initial configuration segment and the initial trimming segment data output by the segment division unit. Both segments are composed of a set of nodes and their spatial connection relationships. The system extracts the boundaries of the two segments and identifies their outer contour node sequences. The boundary identification method includes distance calculation and curvature abrupt change identification for the end edge nodes of each segment. By analyzing the connection angle and normal angle of adjacent nodes, the system finds the locations of boundary folds, discontinuities, or normal abrupt changes.

[0087] After identification, the system performs contour reconstruction on regions with gaps, misalignments, or structural abrupt changes between boundaries. The reconstruction strategy includes two types: for regions with sparse nodes or density mismatches, the system uses interpolation algorithms to automatically generate intermediate interpolation points between boundary segments, thereby filling contour gaps; for regions with abrupt changes in node normals, the system performs smoothing based on the difference between the node's tangent and normal directions, and uses moving averages to adjust the vector direction of the abrupt change points to ensure a continuous transition. After completing boundary reconstruction, the system regenerates connectivity maps for the entire initial and trimmed configuration segments, and corrects the region labels according to the correlation between boundary nodes and the central region, ultimately forming a dataset of configuration and trimmed regions with continuous boundary definitions.

[0088] In a preferred embodiment of the present invention, the edge-forming module includes:

[0089] The contour extraction unit is used to extract contour points of the boundary of the configuration area based on the structural distribution data, and obtain contour point sequence data.

[0090] The gradient analysis unit is used to perform spatial direction change rate analysis on the contour point sequence data, calculate the tangential direction difference between adjacent contour points, and obtain contour gradient data.

[0091] The fitting building unit is used to fit the boundary lines of the contour points based on the contour gradient data to generate the configuration boundary.

[0092] The path generation unit is used to perform equal-layer scanning along the configuration boundary, extract the closed contour lines on each section, generate layer-by-layer processing paths, and obtain configuration path data.

[0093] In this embodiment of the invention, the contour extraction unit is used to extract contour points of the configuration area boundary based on structural distribution data to obtain contour point sequence data. This solves the problems of incomplete contours and disordered point order caused by manual contour extraction in traditional manufacturing processes, laying a geometric data foundation for subsequent contour fitting and path generation. The gradient analysis unit is used to perform spatial direction change rate analysis on the contour point sequence data, calculate the tangential direction difference between adjacent contour points, and obtain contour gradient data, effectively preventing manufacturing problems such as path jumps and uneven printing stacking caused by drastic contour changes. The fitting and construction unit is used to fit the contour points to the boundary lines based on the contour gradient data to generate configuration boundaries, effectively improving the geometric accuracy and descriptive ability of the contour boundaries, and providing a mathematical basis for the interlayer superposition of subsequent configuration paths. The path generation unit is used to perform equal-layer scanning along the configuration boundary, extract closed contour lines on each cross section, generate layer-by-layer processing paths, and obtain configuration path data, ensuring the continuity and spatial consistency of the configuration path between layers, and avoiding problems such as path breakage and layer offset during the printing process.

[0094] The fitting construction unit is used to fit the contour points to the boundary lines based on the contour gradient data, generating the configuration boundary, specifically including:

[0095] The system first performs preprocessing on the contour point sequence, including removing duplicate points, eliminating outliers, and spatial sorting to ensure good fitting continuity and directional consistency. The sorting method is primarily based on the cumulative distance index along the curve length direction, and then adjusted secondaryly according to the direction of spatial angle change to ensure that the fitted input sequence maintains a consistent orientation in space, thereby avoiding loops or interleaving errors in the fitted path.

[0096] The system then automatically detects regions of abrupt directional changes based on the gradient data calculated from the contour point sequence. If the tangential angle between consecutive point segments exceeds a preset gradient threshold (e.g., 10°), it is considered a contour corner or a point of geometric abrupt change. Based on this, the system divides the entire contour into several segment regions, each with gentle directional changes and strong geometric continuity. For these regions, the system uses different fitting methods: cubic splines are used for gentle regions, while B-spline curves are introduced for abrupt changes. During the fitting process, the system performs error constraint checks on each curve segment, controlling the maximum fitting residual to not exceed 10% of the mean distance between the original points. Any excess is corrected through control point redistribution and fitting tension adjustment. After all segmented fittings are completed, the system splices the fitted segments into a closed boundary curve and applies curvature continuity checks and smoothing operations at the curve connections to avoid curvature abrupt changes at transition points, generating configuration boundary data.

[0097] The path generation unit is used to perform equal-layer scanning along the configuration boundary, extract the closed contour lines on each cross section, generate layer-by-layer processing paths, and obtain configuration path data, specifically including:

[0098] First, a unified spatial slicing coordinate system is established. Typically, the Z-axis is chosen as the configuration scanning axis, and a fixed layer height (e.g., 0.2mm or 0.3mm) is set according to the configuration accuracy requirements of the processing equipment, serving as the equidistant slicing plane. The system performs equidistant scanning operations perpendicular to the Z-axis, extracting cross-sectional data layer by layer within the configuration boundary envelope, determining the intersection position of each slicing plane with the configuration boundary, and extracting the closed contour formed by the intersection line.

[0099] The system first voxelizes the fitted boundary to generate an envelope mesh, and then sequentially performs intersection operations with scan surfaces parallel to the XY plane. Intersection points are obtained analytically using a surface-to-plane intersection algorithm and reconstructed into closed curves according to the contour sequence, forming the 2D path contour of the current layer. For path layers with path discontinuities, intersecting lines, or incorrect sequence, the system automatically applies a topology repair algorithm for closure verification, including path endpoint stitching, contour direction consistency adjustment, and isolated segment exclusion. After extracting the closed contour of each layer, the system organizes the inter-layer paths sequentially according to the manufacturing order, forming a complete processing path stack structure. If there are overlapping configurations or paths from different starting layers in the model, the system applies an inter-layer priority control strategy to locally adjust the path execution order, ensuring that each configuration path has coherent manufacturing logic and a stable support structure.

[0100] In a preferred embodiment of the present invention, the mapping module includes:

[0101] The boundary filtering unit is used to extract the boundary segments adjacent to the trimming area based on the spatial distance between the boundary segments and the trimming area, and obtain the set of adjacent boundary segments.

[0102] The projection vector unit is used to construct a normal projection vector for each boundary point in the adjacent boundary segment set, and to filter the direction based on the relationship between the projection direction and the position of the trimmed area to obtain the target projection vector set.

[0103] The spatial mapping unit is used to project the target projection vector set onto the trimmed target surface, establish the spatial correspondence between boundary points and projection points, and form a set of mapping line segments.

[0104] The path construction unit is used to sequentially connect the mapped line segment set to generate a continuous path line structure and obtain preliminary path data.

[0105] In this embodiment of the invention, the boundary filtering unit is used to extract the boundary segments adjacent to the trimming area based on the spatial distance between the boundary segments and the trimming area, thereby obtaining a set of adjacent boundary segments. This accurately extracts the key boundary segments in the configuration boundary that are spatially adjacent to the trimming area, thus providing an effective set of target points for subsequent mapping. The projection vector unit is used to construct a normal projection vector for each boundary point in the set of adjacent boundary segments, and to perform direction filtering based on the relationship between the projection direction and the position of the trimming area, thereby obtaining a set of target projection vectors. This ensures that the direction information used in the subsequent mapping process has a clear physical meaning for processing. The spatial mapping unit is used to project the set of target projection vectors onto the trimming target surface, construct the spatial correspondence between the boundary points and the projection points, form a set of mapping line segments, clarify the positional mapping of the actual deviation of the configuration boundary on the trimming surface, and avoid path breaks or abrupt changes. The path construction unit is used to sequentially connect the set of mapping line segments to generate a continuous path line structure, obtain preliminary trimming path data, and transform the discrete set of mapping points into a continuous path structure, thus completing the conversion from spatial coordinate data to executable path data.

[0106] The projection vector unit is used to construct a normal projection vector for each boundary point in the adjacent boundary segment set, and to perform direction filtering based on the relationship between the projection direction and the position of the trimmed area to obtain the target projection vector set, specifically including:

[0107] The system first extracts the set of pre-selected adjacent boundary points from the configuration boundary and performs local geometric analysis on each boundary point to obtain its surface normal direction. This normal direction can be obtained by analyzing the geometric context of the boundary point in the 3D model (such as the local cross-section or configuration layer formed with adjacent points), and further estimated using vector cross product, differential methods, or point cloud-based normal estimation algorithms. Next, after obtaining the normal vector direction of each boundary point, the system compares it with its spatial relative position in the trimming region. This comparison process is based on the spatial topological information between the configuration region and the trimming region, calculating the angle between the projection vector and the normal of the trimming target surface. If the angle is within a threshold range (e.g., less than 45 degrees), it indicates that the direction has physical contact feasibility and can be used for path projection; otherwise, the boundary point is considered invalid and discarded. By completing the transformation from the boundary point set to the projection vector set, the output structure is an ordered vector set, where each element is a unit normal vector, and a unique index relationship is established between it and its corresponding boundary point, resulting in the target projection vector set.

[0108] The spatial mapping unit is used to project the target projection vector set onto the trimmed target surface, establish the spatial correspondence between boundary points and projection points, and form a set of mapping line segments, specifically including:

[0109] The system first linearly extends each boundary point and its corresponding unit projection vector, projecting it as a ray in the projection direction to form a ray originating from the boundary point. Next, based on the topological definition of the trimming region in the 3D model, the system determines the position, boundary range, and normal direction of the trimming target surface. The system then calculates the intersection point between the projected ray and the trimming target surface, determining whether the ray intersects the target surface within a specified distance (e.g., within 20mm). If a successful intersection occurs, the intersection point is recorded as the spatial projection point of that boundary point; otherwise, the boundary point is marked as a failure and removed. The system establishes a one-to-one mapping relationship between all valid boundary points and their corresponding projection points, constructing a set of mapping line segments. Each line segment consists of a boundary point and its projection point, resulting in a set of mapping line segments.

[0110] In a preferred embodiment of the present invention, the torsion module includes:

[0111] The path segment radian unit is used to extract the included angle between any two adjacent preliminary trimming paths in the preliminary trimming path data, convert it into radian values, remove the units of measurement, and obtain the path radian data.

[0112] The curvature measurement construction unit is used to extract the length of each preliminary trimming path segment and, in combination with the path arc data, calculate the curvature radius of each preliminary trimming path segment to obtain curvature radius data.

[0113] The rate of change analysis unit is used to calculate the rate of change of arc between adjacent preliminary trimming paths based on the path arc data;

[0114] The boundary torsion value unit is used to calculate the boundary torsion value based on the path radii data, radius of curvature data, and rate of change of radii.

[0115] In this embodiment of the invention, the path segment radian unit is used to extract the included angle between any two adjacent preliminary trimmed paths in the preliminary trimmed path data, convert it into radian values, remove the units of measurement, and obtain path radian data, thereby realizing the quantitative characterization of path direction changes and providing basic data support for subsequent path continuity analysis; the curvature quantity construction unit is used to extract the length of each preliminary trimmed path segment, and combine it with the path radian data to calculate the radius of curvature of each preliminary trimmed path segment, thereby obtaining radius of curvature data, realizing the quantitative expression of the smoothness in the path structure, and effectively identifying abrupt bends or abrupt changes in the path; the rate of change analysis unit is used to calculate the rate of change of radian between adjacent preliminary trimmed paths based on the path radian data, and identify the stability of the continuous change trend in the path structure; the boundary torsion value unit is used to calculate the boundary torsion value based on the path radian data, radius of curvature data, and rate of change of radian, reflecting the composite characteristics of the path in terms of spatial morphology and change trend.

[0116] The path segment radian unit is used to extract the angle between any two adjacent preliminary trimming paths in the preliminary trimming path data, convert it into radian values, remove the units, and obtain the path radian data, specifically including:

[0117] Each initial path trimming is considered to consist of multiple consecutive path segments, each segment consisting of two adjacent nodes. To calculate the angle between adjacent path segments, the system first constructs direction vectors for any two consecutive path segments consisting of three points. Specifically, it is assumed that there are three consecutive path nodes. , and Then, the path segment direction vectors are constructed respectively through vector difference operations. and Next, the two direction vectors are normalized to unit vectors to ensure that the calculation results are not affected by the path segment length. After obtaining the unit vectors, the cosine of their included angle is calculated using the vector dot product formula, i.e.: Then, the inverse cosine function is called to obtain the included angle value, and its unit is converted from angle to dimensionless radians. To ensure numerical stability, the system performs boundary clipping on the included angle cosine value before inputting the inverse cosine function, ensuring that it falls within the domain interval [−1, 1] to avoid numerical errors. Finally, the obtained radian value represents the turning strength between two adjacent path segments and is saved by the system as path segment radian data.

[0118] The curvature measurement unit is used to extract the length of each preliminary trimming path segment and, combined with the path radii data, calculate the radius of curvature of each preliminary trimming path segment to obtain the radius of curvature data, specifically including:

[0119] The system first calculates the distance between all adjacent node pairs in the initial path trimming to obtain the spatial length data of each path segment. For any segment in the path, the distance between nodes is calculated. and The resulting path segments are calculated using the Euclidean distance formula, i.e.: ,in Represents path segment Length, This represents the three-dimensional coordinates of the node. The system performs the above operation sequentially on all segments of the path, forming a complete sequence of path segment lengths. Subsequently, the system combines the length sequence with the path segment radii data generated in the previous stage to calculate the radius of curvature for each path segment using the following formula: ,in Represents path segment The radius of curvature, The length of this segment. This represents the radian value corresponding to that segment. When the radian value of a certain path segment approaches 0, the system sets a minimum threshold to avoid numerical instability caused by division by zero. The system can standardize all calculated radii of curvature or remove outliers to obtain the radii of curvature data.

[0120] In a preferred embodiment of the present invention, the boundary torsion value unit includes:

[0121] The boundary torsion value calculation unit is used to extract the direction angle value of each preliminary trimming path segment and fuse it with the path length and the radius of curvature of the corresponding segment to capture high-risk bending paths and obtain a bending density term; fuse the rate of change of radians with the radius of curvature of the corresponding segment to measure the instability in the continuous change process of the path and obtain a direction change rate term; fuse the rate of change of radians with the radius of curvature of the corresponding segment to suppress the drastic amplification caused by small curvature and obtain a curvature suppression term; and obtain an adjustment factor term based on the length ratio of the corresponding segments.

[0122] By fusing the suppression curvature term and the regulation factor term, the control discontinuity caused by abrupt changes between local path segments is identified, resulting in the inter-segment abrupt change modulation term. The bending density term, the normalization term for the rate of change of direction, and the inter-segment abrupt change modulation term are then fused to obtain the boundary torsion value.

[0123] In this embodiment of the invention, the boundary torsion value calculation unit is used to extract the direction angle value of each preliminary trimming path segment and fuse it with the path length and the radius of curvature of the corresponding segment to capture high-risk bending paths and obtain a bending density term, effectively identifying areas in the path where direction changes frequently occur; it fuses the rate of change of arc with the radius of curvature of the corresponding segment to measure the instability during the continuous change of the path and obtains a direction change rate normalization term, identifying abrupt direction changes in the path; it fuses the rate of change of arc with the radius of curvature of the corresponding segment to suppress drastic amplification caused by small curvature and obtains a suppressed curvature term, ensuring that only when the angle changes drastically and the radius of curvature changes simultaneously... The system only identifies dangerous path segments when threshold conditions are met. Based on the length ratio of the corresponding segments, a regulation factor is obtained, which enhances the identification of structural abrupt changes in the path by strengthening the risk identification caused by length abrupt changes. The suppression curvature term is fused with the regulation factor to identify control discontinuities caused by abrupt changes between local path segments, resulting in an inter-segment abrupt change modulation term. This term significantly amplifies the evaluation function response value when small-radius bends and asymmetrical length distributions occur in the path. Finally, the bending density term, the normalization term for the rate of change of direction, and the inter-segment abrupt change modulation term are fused to obtain the boundary torsion value, providing a dynamic control strategy basis for the manufacturing path and serving as data support for the transition from geometric description to numerical decision-making.

[0124] The formula for calculating the boundary torsion value is as follows: ,

[0125] in, The boundary torsion value, This is an index for preliminary path adjustments. This represents the total number of paths that have been initially modified. For the first The initial adjustment of the path's curvature. For the first The length of the initial path trimming section. For the first The radius of curvature of the initial trimmed path. For the first The rate of change of the arc of the preliminary path trimming between the segment and the adjacent segment. , For the first The initial adjustment of the path's curvature. This represents the maximum radius of curvature in the initial trimming path. It is a local minimum constant.

[0126] in, This is the bending density term; The rate of change of direction is normalized to one term; This is an inter-segment abrupt modulation term. To suppress the curvature term, This is the adjustment factor term.

[0127] In a preferred embodiment of the present invention, the disturbance module includes:

[0128] The path matching unit is used to search for path segments with similar structures in the historical sample path database based on the structure and boundary torsion value of the initially modified path, and generate matching index pairs.

[0129] The interferometric overlay unit is used to extract the interferometric offset from historical samples based on the matching index and to perform vector overlay on the preliminary trimming path to obtain the perturbation fusion path.

[0130] The perturbation feature extraction unit is used to identify concentrated perturbation regions on the perturbation fusion path, extract the offset peak value and gradient change value of the path node, and obtain perturbation feature data.

[0131] The interference amplitude unit is used to construct an interference response function based on disturbance characteristic data, reflecting the overall severity of interference along the path and obtaining the interference amplitude index.

[0132] In this embodiment of the invention, the path matching unit is used to search for path segments with similar structures in the historical sample path database based on the structure and boundary torsion value of the preliminary modified path, and generate matching index pairs, thereby realizing feature alignment and similarity search between different path segments; the interference superposition unit is used to extract the interference offset from the historical samples based on the matching index pairs, and perform vector superposition on the preliminary modified path to obtain the perturbation fusion path, which can avoid possible interference problems in advance and realize the evolution of path data from static planning to dynamic adjustment; the perturbation feature extraction unit is used to identify the concentrated perturbation area on the perturbation fusion path, extract the offset peak value and gradient change value of the path node, and obtain perturbation feature data, laying the foundation for the subsequent quantitative evaluation of the interference amplitude; the interference amplitude unit is used to construct the interference response function based on the perturbation feature data, reflect the overall interference severity of the path, obtain the interference amplitude index, and perform a mechanism for global measurement of the path perturbation intensity, enabling the system to quantify the risk level of each path segment at the numerical level.

[0133] The path matching unit, based on the structure and boundary torsion value of the initially modified path, searches the historical sample path database for path segments with similar structures and generates matching index pairs. Specifically, it includes:

[0134] First, the preliminary trimming path data generated by the mapping module in the current manufacturing task is obtained, and combined with the boundary torsion value output by the torsion module, a structural description vector of the preliminary path segment is constructed. This description vector includes, but is not limited to, the following data information: path segment length, direction vector between the start and end points of the path segment, average curvature value of the path segment, location of maximum curvature abrupt change, local directional gradient of each node in the path segment, interval characteristics of boundary torsion value, node distribution density, etc.

[0135] The system then retrieves the path segment feature index table from the pre-set historical sample path database, extracts features from the large number of known path segments stored therein, and constructs the corresponding sample structure vector set. The path segments in this database are all typical interference fragments recorded in the actual manufacturing process in the past, which are closely related to factors such as machining errors, material accumulation, and tool vibration.

[0136] The system then performs matching operations between the initial path structure vector and the sample structure vector set. Matching algorithms can employ methods such as cosine similarity calculation, Euclidean distance comparison, and KL divergence, using a distance function to measure the structural similarity between the current path segment and each sample path segment. In practice, when the similarity exceeds a preset threshold (e.g., similarity perturbation data ≥ perturbation data 0.85 or distance perturbation data ≤ perturbation data 0.2), the path segment is determined to be structurally similar. Finally, the system constructs a matching index pair from all path segment pairs that pass the determination.

[0137] The interferometric overlay unit is used to extract the interferometric offset from historical samples based on the matching index and to perform vector overlay on the preliminary trimming path to obtain the perturbation fusion path, specifically including:

[0138] First, based on the matching index pairs output by the path matching unit, the system searches for the corresponding path segment records in the historical sample path database one by one, and extracts the interference offset data that occurred during the actual processing of that segment. This offset data is generally stored in the form of a three-dimensional vector, recording the displacement changes of each path node relative to the ideal path during actual execution, including: micro-offset values ​​in the X / Y / Z axes, path direction angle offset values, and nonlinear disturbance trends between nodes caused by processing vibrations.

[0139] Subsequently, the system processes the extracted interference offsets to form a path perturbation vector set. To ensure the consistency of data overlay, the historical offset data needs to be scaled and the number of nodes interpolated to maintain a one-to-one correspondence with the node structure of the current preliminary path adjustment. If the structural dimensions are different, the system will use Bezier interpolation to adjust the number of nodes in the historical path to match the length of the current path segment.

[0140] The system then adds a corresponding perturbation vector to each current path node and performs a vector superposition operation. Vector superposition can employ a weight adjustment mechanism, introducing a superposition weight coefficient. The superposition formula is as follows: ,in, For the current path The position vectors of each node. For the corresponding historical path Interference offset vectors of each node, This represents the position vector of the node after perturbation fusion. (Weights are then added.) The value of can be adaptively set based on the boundary torsion value. For example, a higher weight is set when the torsion value is large to enhance the local compensation, while a lower weight is set when the torsion value is small to avoid introducing excessive disturbance. Finally, the disturbance fusion results of all nodes are assembled into a new path segment to obtain the disturbance fusion path.

[0141] In a preferred embodiment of the present invention, the interference amplitude unit includes:

[0142] The interference amplitude calculation unit is used to calculate the mean, standard deviation and gradient change of the path node offset, and then calculate the mean and standard deviation of the gradient change.

[0143] Based on the peak, mean, and standard deviation of the offset of the path nodes, and combined with their normalized relative positions in the path, the influence of a single outlier on the whole is suppressed, resulting in the amplitude peak normalization amplification term; based on the maximum and mean of the gradient change values ​​of the path nodes, the gradient strength of the path nodes in the overall offset is measured, resulting in the gradient strength comprehensive term; based on the standard deviation and mean of the gradient change values ​​of the path nodes, their coefficient of variation is calculated, and based on this, the relative dispersion of the gradient distribution is measured, resulting in the unbalanced modulation term.

[0144] The interference amplitude index is obtained by fusing the amplitude peak normalization amplification term, the gradient intensity comprehensive term, and the unbalanced modulation term.

[0145] In this embodiment of the invention, the interference amplitude calculation unit is used to calculate the mean, standard deviation, and gradient change of the path node offset, and then calculate the mean and standard deviation of the gradient change to measure the severity of local disturbances; based on the peak value, mean, and standard deviation of the path node offset, and combined with its normalized relative position in the path, the influence of a single outlier on the whole is suppressed, resulting in a normalized amplification term for the amplitude peak, effectively suppressing the misleading effect of extreme values ​​on the overall interference index; based on the maximum value and mean of the gradient change value of the path node, the gradient intensity of the path node in the overall offset is measured, resulting in a comprehensive gradient intensity term, emphasizing the abrupt change region of path disturbance. The system's ability to capture data effectively identifies areas in the path where direction changes are rapid or node errors accumulate sharply. Based on the standard deviation and mean of the gradient changes at path nodes, its coefficient of variation is calculated. This coefficient measures the relative dispersion of the gradient distribution, resulting in an unbalanced modulation term. This term measures the discreteness of the gradient distribution under disturbance conditions, revealing whether there are drastic disturbance segments in multiple local regions of the path. By fusing the amplitude peak normalization amplification term, the gradient intensity comprehensive term, and the unbalanced modulation term, an interference amplitude index is obtained. This constructs a multi-dimensional, multi-granular information integration and evaluation model, providing stable and reliable basic data support for subsequent path correction.

[0146] The formula for calculating the interference amplitude index is as follows: ,

[0147] in, The interference magnitude index for the path, This represents the peak offset of the path node. This represents the average offset of the path nodes. The standard deviation of the path node offset. for The normalized relative position of the path node within the path. This represents the maximum value of the gradient change at the path node. This represents the mean of the gradient changes at the path nodes. The coefficient of variation is the gradient change value of the path node. , The standard deviation of the gradient changes at the path nodes. It is a local minimum constant.

[0148] in, For the amplitude peak normalization amplification term, This is the comprehensive term for gradient strength. This is an unbalanced modulation term.

[0149] In a preferred embodiment of the present invention, the correction module includes:

[0150] The weight generation unit is used to allocate the magnitude according to the interference magnitude index, and to proportionally divide the interference magnitude index into a weight sequence corresponding to each path node to obtain weight data.

[0151] The offset calculation unit is used to calculate the coordinate increment based on the weight data and the distance between adjacent path nodes, calculate the offset increment of each path node on the three axes, and obtain the coordinate increment sequence.

[0152] The orientation calculation unit is used to adjust the orientation based on the coordinate increment sequence and the orientation sequence of the preliminary trimming path, generate orientation coefficients corresponding to each path node, and obtain the orientation coefficient sequence.

[0153] The point update unit is used to update the node position and direction vector according to the coordinate increment sequence and the direction coefficient sequence, so as to obtain the updated node set;

[0154] The constraint projection unit is used to perform boundary constraint projection based on the structural distribution data, and to revise the projection of out-of-bounds nodes in the updated node set to obtain a boundary compliant node set.

[0155] The path fitting unit is used to perform continuous curve fitting and inter-segment connection to generate a path segment sequence based on the boundary compliance node set, thereby obtaining the fitted path data.

[0156] In this embodiment of the invention, a weight generation unit is used to allocate the interference amplitude index according to the interference amplitude index, proportionally dividing the interference amplitude index into a weight sequence corresponding to each path node to obtain weight data, thereby realizing spatial distribution control of the disturbance correction process and avoiding average correction of the path; an offset calculation unit is used to calculate the coordinate increment based on the weight data and the distance between adjacent path nodes, calculating the offset increment of each path node on the three axes to obtain a coordinate increment sequence, thereby realizing the mapping of interference effects to specific three-dimensional coordinates; a direction calculation unit is used to adjust the direction based on the coordinate increment sequence and the direction sequence of the preliminary path trimming, generating direction coefficients corresponding to each path node to obtain a direction coefficient sequence, significantly reducing tool runout, path interruption, or movement caused by sudden changes in direction. The system comprises four parts: a point update unit, a point position update unit, and a path curve replanning unit. The point update unit updates the node position and direction vector based on the coordinate increment sequence and direction coefficient sequence to obtain an updated node set. This completes the reconstruction from disturbance feedback to geometric space, achieving the effect of replanning the path curve. The constraint projection unit performs boundary constraint projection based on the structural distribution data, revising the projection of out-of-bounds nodes in the updated node set to obtain a boundary compliant node set. This prevents path data from going out of bounds, inter-layer interference, or structural conflicts caused by corrections, ensuring that all processing trajectories are within the manufacturable area. The path fitting unit performs continuous curve fitting and inter-segment connection to generate a path segment sequence based on the boundary compliant node set, obtaining fitted path data. This transforms the discrete node set into a continuous manufacturable path, improving the smoothness and structural consistency of the path.

[0157] The weight generation unit is used to allocate amplitude based on the interference amplitude index, proportionally dividing the interference amplitude index into a weight sequence corresponding to each path node to obtain weight data, specifically including:

[0158] To refine the interferometric amplitude index into personalized correction criteria for each node, a structural analysis of the path node set is first performed. This extracts the relative position of each node within the path, its local offset peak value, and its gradient change value. These factors are then combined to form the local interferometric intensity factor for each node. After normalizing the local factors of all nodes, a normalized interferometric influence coefficient vector is formed. The interferometric amplitude index is then distributed proportionally according to this vector, generating a weight value corresponding to each path node. This distribution process can employ a linear weight mapping model, where the weight value of each node equals the total interferometric amplitude index multiplied by its corresponding normalization factor, forming a weight sequence.

[0159] The constraint projection unit is used to perform boundary constraint projection based on the structural distribution data, and to revise the projection of out-of-bounds nodes in the updated node set to obtain a boundary-compliant node set, specifically including:

[0160] The system first imports the structural distribution data generated during the partitioning of the configuration and trimming zones. This data defines the geometric extent of the model boundaries, region labels, and boundary point sets. After correcting the node coordinates, some path nodes may exceed the boundary range due to cumulative offset, resulting in path out-of-bounds errors. To prevent manufacturing paths from entering unprocessable areas, the system iterates through and updates the node set sequentially, calculating whether the spatial position of each node falls within the defined structural boundaries. If node coordinates are detected outside the component boundaries, a projection revision process is triggered. This process first constructs the projection direction based on the node's current direction vector, then finds the boundary point or boundary plane with the shortest distance within its direction range, and maps the out-of-bounds node coordinates onto the legal boundary surface through normal vector projection, generating the projected node coordinates. For nodes whose projection points are in the boundary corner area, a position correction threshold mechanism is also added to prevent abnormal local node density caused by clustered projection. To maintain path continuity, the projected node sequence needs to be reordered and the spacing adjusted to ensure that the distance between nodes is controlled within the preset tolerance range. The final node set is the boundary-compliant node set.

[0161] The path fitting unit is used to perform continuous curve fitting and inter-segment connection to generate a path segment sequence based on the boundary compliance node set, thereby obtaining the fitted path data, specifically including:

[0162] First, the curve fitting strategy is determined based on the spatial distribution of the node sequence. If the nodes are evenly distributed, a third-order B-spline curve algorithm is preferred; if the node distribution is dense or there are inflection points, a Bezier curve is used. The fitting process includes two stages: curve construction and inter-segment connection. In the curve construction stage, the system uses the node sequence as control points and the direction vectors of the nodes as tangent constraints to perform local continuity fitting, generating a series of high-order continuous curve segments. Each curve segment maintains at least one continuous derivative to ensure curvature stability. In the inter-segment connection stage, to avoid abrupt path changes, the system introduces a transition segment generation mechanism. Transition segments are constructed based on the difference between the endpoint and starting directions of adjacent curve segments, and interpolation algorithms are used for transition connections, forming a smooth, non-jumping overall path structure. After fitting, the system organizes the path segment sequence into a structured data format, including the start and end points, parameter functions, and direction vectors of each path segment, ultimately generating the fitted path data.

[0163] In a preferred embodiment of the present invention, the splicing module includes:

[0164] The sequential parsing unit is used to sequentially parse the fitted path data and configuration path data based on the timing information of the manufacturing process to obtain the path execution sequence data.

[0165] The endpoint matching unit is used to extract the start and end nodes of the fitted path data and the configuration path data based on the path execution order data, perform endpoint distance calculation and shortest connection matching, and obtain endpoint matching pairs.

[0166] The connection generation unit is used to splice path segments according to endpoint matching pairs, and insert transition segments at the splicing points to ensure curvature continuity, thereby obtaining mixed path data;

[0167] The format unification unit is used to convert mixed path data into a unified path data structure format, perform node index rearrangement and segment attribute labeling to obtain structured mixed path data, and translate it into a control instruction set that can be recognized by the manufacturing equipment to obtain mixed execution path instructions.

[0168] In this embodiment of the invention, the sequential parsing unit is used to sequentially parse the fitted path data and the configuration path data according to the timing information of the manufacturing process flow to obtain the path execution sequence data. This realizes the systematic and procedural processing of the path data execution sequence, avoiding the manufacturing process disorder caused by the disorder of path segment sequence. The endpoint matching unit is used to extract the start and end nodes of the fitted path data and the configuration path data according to the path execution sequence data, perform endpoint distance calculation and shortest connection matching to obtain endpoint matching pairs, ensuring that the splicing point has high spatial continuity and directional coordination, and avoiding tool failures caused by misconnected paths or directional conflicts. Trajectory jumps or position offsets; connection generation unit, used to splice path segments according to endpoint matching pairs, and insert transition segments at the splicing points to ensure curvature continuity, to obtain mixed path data, achieving natural transitions between path segments and avoiding vibrations or sudden changes during machine tool execution; format unification unit, used to convert mixed path data into a unified path data structure format, perform node index rearrangement and segment attribute labeling, to obtain structured mixed path data, and translate it into a control instruction set that manufacturing equipment can recognize, to obtain mixed execution path instructions, ensuring that path data achieves standardization and compatibility in three dimensions: structural layer, semantic layer, and instruction layer.

[0169] The connection generation unit is used to splice path segments based on endpoint matching pairs, and insert transition segments at the splicing points to ensure curvature continuity, thereby obtaining mixed path data. Specifically, it includes:

[0170] First, based on the position coordinates and tangential direction information of the starting and ending nodes of the path segment, the initial geometric conditions of the connecting segment are constructed. The system uses the end point of the configuration path and the starting point of the fitted path as endpoints to establish a set of interpolation reference points in three-dimensional space. Then, directional constraints are constructed based on the tangent directions of the two paths near the connecting point, thereby limiting the starting and ending derivative directions of the connecting path. Subsequently, the system uses a spline interpolation algorithm to fit the connecting segment, generating a transition path that satisfies the conditions of continuous first derivative and finite second derivative. To avoid abrupt changes in the spatial curvature of the connecting segment that could lead to trajectory discontinuities or tool impacts during equipment execution, the system introduces a minimum radius of curvature constraint during the fitting process. This constraint controls the maximum gradient of curvature change per unit length of the path, preventing the formation of sharp corners.

[0171] After curve fitting, the system discretely samples the path points of the connecting segments using a constant step size, generating a node sequence. The sampling interval is adaptively set based on the required manufacturing precision and equipment resolution. The system then assigns attribute tags to the connecting segment, defining it as a connecting segment type, and inserts it between the end point of the configuration path segment and the start point of the trimming path segment, thus forming a structurally continuous sequence of hybrid path segments. To ensure smooth path execution, the system also verifies the tangential angle between the connecting segment and the preceding and following path segments. If a sudden change in direction between consecutive segments is found to exceed a preset threshold, the connecting segment's directional constraints are recalculated until a stable transition is achieved. Finally, the system integrates all the interpolated and attribute-calibrated connecting segments into a unified path segment set to form hybrid path data.

[0172] The format unification unit is used to convert mixed path data into a unified path data structure format, perform node index rearrangement and segment attribute labeling to obtain structured mixed path data, and translate it into a control instruction set that the manufacturing equipment can recognize to obtain mixed execution path instructions, specifically including:

[0173] First, the path segments in the path segment set are sorted according to their order in the manufacturing process, including configuration path segments, connection path segments, and fitting and trimming path segments. The sorting is based on the execution sequence data and the master index matching to ensure the time consistency of the path segments during execution. After sorting, the system uniformly numbers all nodes within the path segments, regenerates the node index table, and establishes a globally unique index mapping relationship for the nodes.

[0174] After node indexing is completed, the system enters the path segment attribute labeling stage. Each path segment is assigned attribute tags based on its generation source, including segment type (configuration segment, connection segment, trimming segment), processing method (additive, subtractive, transition), execution priority, etc. Attribute tags are attached to the path segment's metadata as structured fields, giving the path data complete process semantics during transmission and parsing. Subsequently, the system formats and transforms the data structure of all path segments, uniformly encapsulating information such as node sequences, direction vectors, speed suggestions, processing modes, and hierarchical relationships into a data structure format conforming to preset standards, such as using nested JSON structures or XML structures supported by the device.

[0175] After structuring the data format, the system converts the mixed path data into a set of control instructions recognizable by the manufacturing equipment according to the manufacturing equipment control protocol. This includes operations such as coordinate format conversion, interpolation mode setting, speed command synthesis, and tool path call format matching. For example, for the G-code control system, the system converts path points into a set of G1 / G2 / G3 linear or circular interpolation instructions, adds corresponding parameters such as F value (feed rate), S value (spindle speed), and T value (tool number), and adds safety movement instructions and coordinate reset instructions at the start and end of the path to generate mixed execution path instructions.

[0176] The above description represents the preferred embodiments of the present invention. It should be noted that those skilled in the art can make various improvements and modifications without departing from the principles of the present invention, and these improvements and modifications should also be considered within the scope of protection of the present invention.

Claims

1. A hybrid manufacturing execution system combining precision machining and 3D printing, characterized in that, The system includes: The partitioning module is used to identify regions based on the 3D model of the target component, dividing the 3D model into configuration regions and trimming regions to obtain structural distribution data; The edge-forming module is used to perform gradient fitting on continuous contour points within the configuration area based on structural distribution data, generate configuration boundaries, and generate layer-by-layer closed contour paths along the configuration boundaries to obtain configuration path data. The mapping module is used to extract boundary segments close to the trimming area based on the configuration boundary data, and calculate their spatial projection vectors on the trimming target surface to obtain preliminary trimming path data. The torsion module extracts the angle between any two adjacent preliminary trimming paths from the preliminary trimming path data, converts it to radians, removes the units, and obtains the path radian data. It also extracts the length of each preliminary trimming path segment and, combined with the path radian data, calculates the radius of curvature for each segment, obtaining the radius of curvature data. Furthermore, it calculates the rate of change of radians between adjacent preliminary trimming paths based on the radian data. By extracting the direction angle value of each preliminary trimming path segment and fusing it with the path length and the corresponding segment's radius of curvature, it identifies high-risk bending paths and obtains the bending density term. The radian change rate is fused with the radius of curvature of the corresponding segment to measure the instability during the continuous path change process, resulting in a normalized term for the directional change rate. The radian change rate is also fused with the radius of curvature of the corresponding segment to suppress the drastic amplification caused by small curvature, resulting in a suppressed curvature term. A modulating factor term is obtained based on the length ratio of the corresponding segments. The suppressed curvature term and the modulating factor term are fused to identify the control discontinuity caused by abrupt changes between local path segments, resulting in an inter-segment abrupt change modulation term. Finally, the bending density term, the normalized term for the directional change rate, and the inter-segment abrupt change modulation term are fused to obtain the boundary torsion value. The perturbation module is used to search for path segments with similar structures in the historical sample path database based on the structure and boundary torsion value of the initial trimmed path, generating matching index pairs; extracting the interference offset from the historical samples based on the matching index pairs, and performing vector superposition on the initial trimmed path to obtain the perturbation fusion path; identifying the perturbation concentration region on the perturbation fusion path, extracting the offset peak value and gradient change value of the path nodes to obtain perturbation feature data; calculating the mean, standard deviation, and gradient change of the path node offsets, and then calculating the mean and standard deviation of the gradient change; based on the path nodes... The peak, mean, and standard deviation of the offset are calculated, and combined with their normalized relative positions in the path, to suppress the influence of individual outliers on the whole, resulting in an amplitude peak normalization amplification term. Based on the maximum and mean values ​​of the gradient changes of the path nodes, the gradient strength of the path nodes in the overall offset is measured, resulting in a gradient strength comprehensive term. Based on the standard deviation and mean values ​​of the gradient changes of the path nodes, their coefficient of variation is calculated, and based on this, the relative dispersion of the gradient distribution is measured, resulting in an unbalanced modulation term. The amplitude peak normalization amplification term, the gradient strength comprehensive term, and the unbalanced modulation term are then fused to obtain the interference amplitude index. The correction module is used to correct the coordinates of the preliminary correction path according to the interference amplitude index, adjust the position and direction of each preliminary correction path point, and obtain the fitted path data. The splicing module is used to connect and unify the paths according to the manufacturing sequence based on the fitted path data and the configuration path data to obtain mixed execution path instructions.

2. The precision machining and 3D printing hybrid manufacturing execution system according to claim 1, characterized in that, The partitioning module includes: The node extraction unit is used to extract nodes from the 3D model and perform geometric distribution sampling to obtain discrete node data. The curvature calculation unit is used to measure the local surface curvature of the three-dimensional model based on discrete node data to obtain curvature distribution data. The segmentation unit is used to perform preliminary region classification of the 3D model based on curvature distribution data, forming the initial configuration segment and the initial trimming segment; Boundary fusion units are used to reorganize the contour lines between the initial configuration segment and the initial trimming segment, adjust the node distribution of discontinuous boundaries, form configuration and trimming regions, and obtain structural distribution data.

3. The precision machining and 3D printing hybrid manufacturing execution system according to claim 2, characterized in that, The edge-forming module includes: The contour extraction unit is used to extract contour points of the boundary of the configuration area based on the structural distribution data, and obtain contour point sequence data. The gradient analysis unit is used to perform spatial direction change rate analysis on the contour point sequence data, calculate the tangential direction difference between adjacent contour points, and obtain contour gradient data. The fitting building unit is used to fit the boundary lines of the contour points based on the contour gradient data to generate the configuration boundary. The path generation unit is used to perform equal-layer scanning along the configuration boundary, extract the closed contour lines on each section, generate layer-by-layer processing paths, and obtain configuration path data.

4. The precision machining and 3D printing hybrid manufacturing execution system according to claim 3, characterized in that, The mapping module includes: The boundary filtering unit is used to extract the boundary segments adjacent to the trimming area based on the spatial distance between the boundary segments and the trimming area, and obtain the set of adjacent boundary segments. The projection vector unit is used to construct a normal projection vector for each boundary point in the adjacent boundary segment set, and to filter the direction based on the relationship between the projection direction and the position of the trimmed area to obtain the target projection vector set. The spatial mapping unit is used to project the target projection vector set onto the trimmed target surface, establish the spatial correspondence between boundary points and projection points, and form a set of mapping line segments. The path construction unit is used to sequentially connect the mapped line segment set to generate a continuous path line structure and obtain preliminary path data.

5. The precision machining and 3D printing hybrid manufacturing execution system according to claim 4, characterized in that, The correction module includes: The weight generation unit is used to allocate the magnitude according to the interference magnitude index, and to proportionally divide the interference magnitude index into a weight sequence corresponding to each path node to obtain weight data. The offset calculation unit is used to calculate the coordinate increment based on the weight data and the distance between adjacent path nodes, calculate the offset increment of each path node on the three axes, and obtain the coordinate increment sequence. The orientation calculation unit is used to adjust the orientation based on the coordinate increment sequence and the orientation sequence of the preliminary trimming path, generate orientation coefficients corresponding to each path node, and obtain the orientation coefficient sequence. The point update unit is used to update the node position and direction vector according to the coordinate increment sequence and the direction coefficient sequence, so as to obtain the updated node set; The constraint projection unit is used to perform boundary constraint projection based on the structural distribution data, and to revise the projection of out-of-bounds nodes in the updated node set to obtain a boundary compliant node set. The path fitting unit is used to perform continuous curve fitting and inter-segment connection to generate a path segment sequence based on the boundary compliance node set, thereby obtaining the fitted path data.

6. The precision machining and 3D printing hybrid manufacturing execution system according to claim 5, characterized in that, The splicing module includes: The sequential parsing unit is used to sequentially parse the fitted path data and configuration path data based on the timing information of the manufacturing process to obtain the path execution sequence data. The endpoint matching unit is used to extract the start and end nodes of the fitted path data and the configuration path data based on the path execution order data, perform endpoint distance calculation and shortest connection matching, and obtain endpoint matching pairs. The connection generation unit is used to splice path segments according to endpoint matching pairs, and insert transition segments at the splicing points to ensure curvature continuity, thereby obtaining mixed path data; The format unification unit is used to convert mixed path data into a unified path data structure format, perform node index rearrangement and segment attribute labeling to obtain structured mixed path data, and translate it into a control instruction set that can be recognized by the manufacturing equipment to obtain mixed execution path instructions.

Images