Intelligent planning and optimization method and system for injection mold machining path

Abstract

This invention provides an intelligent planning and optimization method and system for injection mold machining paths, belonging to the field of intelligent machining technology. The method includes acquiring a three-dimensional geometric model of the cavity to be machined and process constraints; extracting geometric topology information based on surface features and performing partitioned processing; constructing a multi-objective optimization model based on local curvature distribution characteristics and material removal constraints; dynamically adjusting the connection path generation strategy through adaptive iterative solution of the model to generate a tool trajectory planning scheme; converting the scheme into CNC machining instructions and executing them, while simultaneously collecting actual cutting force feedback information to correct model parameters, thus achieving intelligent machining path optimization. This invention improves machining efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to intelligent manufacturing technology, and more particularly to a method and system for intelligent planning and optimization of injection mold processing paths. Background Technology

[0002] In the field of injection mold manufacturing, CNC machining of the cavity is a crucial step in achieving precision mold forming. Currently, the machining path planning for mold cavities typically relies on computer-aided manufacturing software. Operators manually divide the machining area and generate toolpaths based on the 3D model of the cavity, either manually or using built-in general algorithms. Conventional practices generally employ a fixed partitioning strategy based on geometric features, such as dividing the cavity according to steep and flat areas, and using isoparametric lines or parallel sections to generate toolpaths. Machining parameters, such as the entry point, path spacing, and cutting direction, are often selected based on empirical formulas or recommended values ​​from a process database. The generated toolpath code is then transmitted to the CNC machine tool for execution.

[0003] Fixed-rule partitioning and trajectory planning strategies lack a synergistic consideration of the local geometric features of specific cavities and the constraints of the overall machining process. This can easily lead to unreasonable connection paths between machining zones, resulting in unnecessary idle travel or abrupt tool direction changes in complex surface transition areas. This not only reduces machining efficiency but can also affect surface quality or even damage the tool due to sudden changes in cutting load. Conventional methods often use static preset values ​​for machining parameters, which are difficult to adapt to changes in actual cutting conditions caused by slight variations in material properties and tool wear during machining. The lack of a dynamic adjustment mechanism based on real-time machining feedback causes the process model to become disconnected from actual working conditions, hindering further improvements in machining stability and accuracy consistency. Summary of the Invention

[0004] The present invention provides an intelligent planning and optimization method and system for injection mold processing paths, which can solve the problems in the prior art.

[0005] A first aspect of this invention provides an intelligent planning and optimization method for injection mold processing paths, comprising: Obtain the three-dimensional geometric model of the cavity to be processed and the processing constraints. Extract the geometric topology information of the cavity region based on the surface features of the three-dimensional geometric model. Then, partition the geometric topology information according to the processing constraints to obtain the cavity processing partition with processing priority identifier. Based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints, a multi-objective optimization model for tool trajectory planning is constructed. The multi-objective optimization model is solved by adaptive iteration. During the iteration process, the connection path generation strategy between adjacent machining zones is dynamically adjusted based on the tool trajectory simulation results of the previous iteration, and a tool trajectory planning scheme including the entry point position, cutting direction and line spacing parameters is generated. The tool path planning scheme is converted into CNC machining instructions and output to the machining equipment. At the same time, the actual cutting force feedback information during the machining process is collected, and the material removal constraint parameters in the multi-objective optimization model are corrected using the actual cutting force feedback information.

[0006] Based on the surface features of the three-dimensional geometric model, the geometric topology information of the cavity region is extracted, and the geometric topology information is partitioned according to the processing constraints to obtain cavity processing partitions with processing priority identifiers, including: The three-dimensional geometric model is subjected to surface differential geometry analysis to calculate the principal curvature and Gaussian curvature of each surface unit in the cavity region. The surface feature boundary is identified based on the curvature continuity condition, and geometric topological information containing surface type identifier and boundary topological relationship is extracted. Based on the surface type identifier in the geometric topology information, and combined with the tool accessibility condition and machining direction restriction in the machining process constraints, the cavity region is spatially decomposed, and adjacent surface units with the same machining direction feasibility are merged into the same machining partition. Based on the spatial relationship between the processing partitions and the boundary topology, the processing dependency relationship between each processing partition is determined, and a processing priority identifier is assigned to each processing partition according to the processing dependency relationship, thus obtaining cavity processing partitions with processing priority identifiers.

[0007] The three-dimensional geometric model is subjected to surface differential geometry analysis to calculate the principal curvature and Gaussian curvature of each surface element in the cavity region, and the surface feature boundaries are identified based on the curvature continuity condition, including: The cavity region of the three-dimensional geometric model is discretized into a surface mesh, dividing the cavity region into multiple surface units, and a local parametric coordinate system is established for each surface unit. In the local parameter coordinate system, the first basic form coefficient and the second basic form coefficient of each surface element are calculated, and the principal curvature of the surface element is solved using the first basic form coefficient and the second basic form coefficient. The Gaussian curvature of each surface element is calculated based on the principal curvature, and the principal curvature and the Gaussian curvature are stored as curvature feature attributes of the surface element. Traverse the common boundary between adjacent surface elements, extract the principal curvature of the surface elements on both sides of the common boundary, and calculate the gradient of the principal curvature change of the surface elements on both sides of the common boundary. The principal curvature change gradient is compared with a preset curvature continuity threshold. When the principal curvature change gradient exceeds the curvature continuity threshold, the common boundary is determined to be a surface feature boundary.

[0008] Based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints, a multi-objective optimization model for tool trajectory planning is constructed, including: For each machining zone in the cavity machining zone, the curvature extrema and curvature change gradient within the machining zone are statistically analyzed to form a local curvature distribution feature characterizing the geometric complexity of the machining zone; Based on the local curvature distribution characteristics, the tool contact area morphology characteristics of the machining zone are determined, and the tool path length objective function and the machining surface quality uniformity objective function are established according to the tool contact area morphology characteristics. Obtain the material property parameters and tool geometry parameters in the machining process constraints, construct the material removal volume constraint per unit time based on the material property parameters and tool geometry parameters, and use the material removal volume constraint per unit time as the material removal constraint; Based on the material removal constraints, the tool interference avoidance space boundary and residual height control conditions are determined, and the tool interference avoidance space boundary and residual height control conditions are used as constraint conditions. By integrating the objective function of tool path length, the objective function of surface quality uniformity, and the constraints, a multi-objective optimization model with tool trajectory parameters as decision variables is constructed.

[0009] The multi-objective optimization model is solved through adaptive iteration. During the iteration process, the path generation strategy between adjacent machining zones is dynamically adjusted based on the tool trajectory simulation results of the previous iteration. This generates a tool trajectory planning scheme that includes parameters such as the entry point position, cutting direction, and line spacing. Initialize the decision variables of the multi-objective optimization model, iteratively solve the multi-objective optimization model to obtain an initial tool trajectory candidate scheme; perform tool motion simulation on the initial tool trajectory candidate scheme, calculate the motion trajectory of the tool in each machining zone, and obtain the tool trajectory simulation results of the previous iteration; Extract the tool posture continuity deviation from the tool trajectory simulation results, compare the tool posture continuity deviation with a preset continuity threshold, and when the tool posture continuity deviation exceeds the preset continuity threshold, determine that there is a need for connection path adjustment between adjacent machining zones; For adjacent machining partitions that have the required connection path adjustment, the residual error of the machining surface at the corresponding boundary position in the tool trajectory simulation result is extracted, the distribution characteristics of the residual error of the machining surface are analyzed, the connection path generation strategy between the adjacent machining partitions is dynamically adjusted according to the distribution characteristics, and the row spacing parameter of the adjacent machining partitions is adjusted based on the tool posture continuity deviation. The adjusted connection path generation strategy and the row spacing parameter are updated to the decision variables. Iterative solution and simulation steps are repeated until the tool posture continuity deviation and the residual error of the machining surface meet the preset convergence conditions, and the tool trajectory planning scheme is output.

[0010] The initial tool trajectory candidate scheme is simulated to calculate the tool's motion trajectory in each machining zone. The simulation results of the previous iteration of the tool trajectory are obtained as follows: Extract the tool path node coordinate sequence of each machining zone in the initial tool path candidate scheme, and construct the tool center point motion trajectory based on the tool path node coordinate sequence; Based on the motion trajectory of the tool center point and combined with the tool's geometric parameters, the tool attitude vector at each path node position is calculated. Based on the tool posture vector, calculate the tool posture change between adjacent path nodes, and record the value of the tool posture change at the boundary position of adjacent machining zones as the tool posture continuity deviation. Based on the tool center point motion trajectory and the tool attitude vector, a tool sweep body model is established. Boolean operation is performed on the tool sweep body model to calculate the cavity surface morphology after the tool sweep body model is removed. The deviation value between the cavity surface morphology and the theoretically designed surface is extracted, and the deviation value is used as the residual error of the machining surface. The tool posture continuity deviation and the residual error of the machined surface are combined to form the tool trajectory simulation results of the previous iteration.

[0011] The tool path planning scheme is converted into CNC machining instructions and output to the machining equipment. Simultaneously, actual cutting force feedback information during the machining process is collected. The material removal constraint parameters in the multi-objective optimization model are then corrected using this actual cutting force feedback information, including: The tool path planning scheme is analyzed, including the entry point position, cutting direction, and line spacing parameters. Numerical control (NC) machining instructions are generated based on the entry point position, cutting direction, and line spacing parameters. The NC machining instructions are then output to the machining equipment to control the machining equipment to perform cavity machining operations. During the machining operation, a force sensor collects cutting force data generated when the tool contacts the workpiece in real time. The cutting force data includes the cutting force amplitude and the cutting force fluctuation characteristics. The cutting force data is subjected to time series analysis to extract the average value of the cutting force amplitude and the peak value of the cutting force fluctuation characteristics. The average value and the peak value are used as actual cutting force feedback information. A preset theoretical cutting force is obtained from the multi-objective optimization model. The actual cutting force feedback information is compared with the theoretical cutting force, and the deviation between the actual cutting force feedback information and the theoretical cutting force is calculated. Based on the deviation, the material removal constraint parameters in the multi-objective optimization model are corrected.

[0012] A second aspect of the present invention provides an intelligent planning and optimization system for injection mold processing paths, comprising: The cavity partitioning unit is used to obtain the three-dimensional geometric model of the cavity to be processed and the processing constraints. Based on the surface features of the three-dimensional geometric model, the geometric topology information of the cavity region is extracted, and the geometric topology information is partitioned according to the processing constraints to obtain the cavity processing partition with processing priority identifier. The trajectory optimization unit is used to construct a multi-objective optimization model for tool trajectory planning based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints. The iterative solution unit is used to solve the multi-objective optimization model through adaptive iteration. During the iteration process, it dynamically adjusts the connection path generation strategy between adjacent machining zones based on the tool trajectory simulation results of the previous iteration, and generates a tool trajectory planning scheme that includes the entry point position, cutting direction and line spacing parameters. The instruction output unit is used to convert the tool trajectory planning scheme into CNC machining instructions and output them to the machining equipment. At the same time, it collects the actual cutting force feedback information during the machining process and uses the actual cutting force feedback information to correct the material removal constraint parameters in the multi-objective optimization model.

[0013] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0015] The beneficial effects of this application are as follows: This method significantly improves the accuracy and efficiency of injection mold cavity machining. Geometric topology information is extracted from the surface features of a 3D geometric model and partitioned according to machining process constraints. This allows for a rational planning of the machining sequence within the cavity area, avoiding wasted idle strokes caused by frequent region jumps in traditional machining and effectively shortening the overall machining time. The partitioned structure with machining priority indicators provides a clear basis for subsequent toolpath optimization, ensuring the systematic and coherent nature of the machining process.

[0016] A multi-objective optimization model was constructed based on local curvature distribution and material removal constraints, achieving a balance between several key indicators such as machining efficiency, surface quality, and tool wear in the tool path. This model comprehensively considers the geometric complexity of the cavity and can generate tool paths that better fit the surface changes, thereby reducing overcutting or undercutting caused by improper path planning and directly improving the contour accuracy and surface finish of the machined surface. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating the intelligent planning and optimization method for injection mold processing paths according to an embodiment of the present invention. Figure 2 This is a flowchart illustrating the intelligent identification and optimization process for the curved surface partitioning of the injection mold cavity in an embodiment of the present invention. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0019] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0020] Figure 1 This is a flowchart illustrating the intelligent planning and optimization method for injection mold processing paths according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes: Obtain the three-dimensional geometric model of the cavity to be processed and the processing constraints. Extract the geometric topology information of the cavity region based on the surface features of the three-dimensional geometric model. Then, partition the geometric topology information according to the processing constraints to obtain the cavity processing partition with processing priority identifier. Based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints, a multi-objective optimization model for tool trajectory planning is constructed. The multi-objective optimization model is solved by adaptive iteration. During the iteration process, the connection path generation strategy between adjacent machining zones is dynamically adjusted based on the tool trajectory simulation results of the previous iteration, and a tool trajectory planning scheme including the entry point position, cutting direction and line spacing parameters is generated. The tool path planning scheme is converted into CNC machining instructions and output to the machining equipment. At the same time, the actual cutting force feedback information during the machining process is collected, and the material removal constraint parameters in the multi-objective optimization model are corrected using the actual cutting force feedback information.

[0021] In one optional implementation, the geometric topology information of the cavity region is extracted based on the surface features of the three-dimensional geometric model, and the geometric topology information is partitioned according to the processing constraints to obtain cavity processing partitions with processing priority identifiers, including: The three-dimensional geometric model is subjected to surface differential geometry analysis to calculate the principal curvature and Gaussian curvature of each surface unit in the cavity region. The surface feature boundary is identified based on the curvature continuity condition, and geometric topological information containing surface type identifier and boundary topological relationship is extracted. Based on the surface type identifier in the geometric topology information, and combined with the tool accessibility condition and machining direction restriction in the machining process constraints, the cavity region is spatially decomposed, and adjacent surface units with the same machining direction feasibility are merged into the same machining partition. Based on the spatial relationship between the processing partitions and the boundary topology, the processing dependency relationship between each processing partition is determined, and a processing priority identifier is assigned to each processing partition according to the processing dependency relationship, thus obtaining cavity processing partitions with processing priority identifiers.

[0022] like Figure 2 As shown, the method includes: After obtaining the 3D geometric model of the part, a detailed geometric topology analysis and partitioning of the cavity region are required. For the surface representation of the 3D geometric model, differential geometry methods are used to analyze each surface element of the cavity region one by one. Specifically, for any point on the cavity surface, the first and second fundamental forms at that point are calculated, and two principal curvatures k_1 and k_2 are obtained by solving the surface curvature equation. The principal curvatures reflect the degree of curvature of the surface at that point along two orthogonal principal directions, where the larger principal curvature k_1 corresponds to the direction of the most severe curvature, and the smaller principal curvature k_2 corresponds to the direction of the gentlest curvature. Based on obtaining the principal curvatures, the Gaussian curvature is calculated. With mean curvature The positive or negative property of Gaussian curvature can directly determine the local morphological characteristics of a surface: when the Gaussian curvature is positive, it indicates that the region is an elliptical surface, exhibiting a convex or concave shape; when the Gaussian curvature is negative, it indicates that the region is a hyperbolic surface, exhibiting a saddle shape; when the Gaussian curvature is close to zero, it indicates that the region is a parabolic or planar region.

[0023] To identify the feature boundaries of the cavity surface, the continuity of principal curvature values ​​is detected along the directions of various surface parameters. When the change in principal curvature value between adjacent surface units exceeds a preset threshold, or the sign of the Gaussian curvature changes, a surface feature boundary is determined to exist at that location. Feature boundaries typically correspond to transition points between surface types, such as the boundary between a plane and a cylinder, or the transition between a cylinder and a sphere. Through the identification of feature boundaries, the cavity surface is decomposed into several surface units with distinct geometric attributes. Each surface unit is assigned a surface type identifier, including basic types such as plane, cylinder, cone, sphere, torus, and freeform surface. Simultaneously, the boundary topological relationships between each surface unit are recorded, including the geometry of the boundary curves, the connection method between adjacent surfaces, and the curvature continuity level at the boundaries. This information collectively constitutes the complete geometric topological information of the cavity region.

[0024] After obtaining the geometric topology information, the cavity region is divided into machining zones based on machining process constraints. These constraints mainly involve tool accessibility limitations and machining direction constraints. Tool accessibility analysis needs to consider the tool's geometric dimensions, tool holder length, and the machine tool's degrees of freedom. For each surface element in the cavity, the interference between the tool and surrounding walls when entering from various directions is analyzed. Specifically, a three-dimensional envelope model of the tool is constructed to simulate the process of the tool approaching the surface element along different directions, and the geometric interference between the tool envelope and other areas of the cavity is detected. Machining directions with interference are marked as infeasible directions; machining directions without interference are recorded as the set of feasible machining directions for that surface element.

[0025] Machining direction limitations are also influenced by the machine tool's axis configuration. For three-axis CNC machine tools, the tool axis direction is fixed, typically vertical; therefore, the angle between the normal vector of the surface element and the vertical direction must be within an acceptable range. For five-axis CNC machine tools, the tool can be oriented in any direction, but the range of motion of the rotary axis still needs to be considered. By analyzing the normal vector distribution of each surface element and the machine tool's motion capabilities, the suitable machining direction range for each surface element is determined. Adjacent surface elements with the same or similar machining direction feasibility are merged. To determine if two surface elements are adjacent, based on the boundary topology relationships recorded in the geometric topology information, they are checked to see if they share a common boundary curve. To determine if two surface elements have the same machining direction feasibility, their sets of feasible machining directions are compared; when the intersection ratio of the two sets exceeds a set threshold, they are considered to have the same machining direction feasibility.

[0026] Based on the above rules, a region growing algorithm is used to partition the machining area. An initial surface unit is selected as a seed region, and adjacent surface units with the same machining direction feasibility are searched. Surface units that meet the criteria are included in the same machining area. The same search and merging operation is performed on newly included surface units until no new adjacent surface units that meet the criteria can be found. After completing the partitioning of a machining area, a new seed region is selected from the remaining unpartitioned surface units, and the above process is repeated until all surface units are assigned to a machining area. In this way, the cavity area is decomposed into several machining areas, and the surface units in each machining area have consistent machining direction attributes, allowing the use of the same toolpath planning strategy.

[0027] After completing the spatial decomposition of the cavity area, it is necessary to determine the machining sequence between each machining section, analyze the spatial relationship between each machining section, and determine their hierarchical relationship in the cavity depth direction. For machining sections located at the bottom of the cavity, their completion will not affect the accessibility of other sections; for machining sections located at the top or sidewall of the cavity, their machining hinders the tool's approach to the bottom section. Combining boundary topology relationships, the occlusion relationships between each machining section are identified. When a machining section is located on the tool path of another machining section, the former forms a machining dependency relationship with the latter, that is, the former must be completed before the latter.

[0028] Based on the processing dependencies among all processing partitions, a directed acyclic graph (DAG) is constructed, where each node represents a processing partition and directed edges represent the direction of processing dependencies. This DAG is then topologically sorted to obtain a sequence of processing orders that satisfies all dependencies. Based on the topological sorting result, each processing partition is assigned a processing priority identifier, with partitions ranked higher having higher priority values. Priority identifiers can be integers; smaller values ​​indicate higher processing priority and should be prioritized in the processing plan. This method yields cavity processing partitions with clearly defined processing priority identifiers, providing structured input data for subsequent processing path planning and CNC program generation.

[0029] In one optional implementation, surface differential geometry analysis is performed on the three-dimensional geometric model to calculate the principal curvature and Gaussian curvature of each surface unit in the cavity region, and the surface feature boundaries are identified based on the curvature continuity condition, including: The cavity region of the three-dimensional geometric model is discretized into a surface mesh, dividing the cavity region into multiple surface units, and a local parametric coordinate system is established for each surface unit. In the local parameter coordinate system, the first basic form coefficient and the second basic form coefficient of each surface element are calculated, and the principal curvature of the surface element is solved using the first basic form coefficient and the second basic form coefficient. The Gaussian curvature of each surface element is calculated based on the principal curvature, and the principal curvature and the Gaussian curvature are stored as curvature feature attributes of the surface element. Traverse the common boundary between adjacent surface elements, extract the principal curvature of the surface elements on both sides of the common boundary, and calculate the gradient of the principal curvature change of the surface elements on both sides of the common boundary. The principal curvature change gradient is compared with a preset curvature continuity threshold. When the principal curvature change gradient exceeds the curvature continuity threshold, the common boundary is determined to be a surface feature boundary.

[0030] After obtaining the 3D geometric model of the cavity, the cavity region is discretized using a surface mesh. Triangular or quadrilateral meshes are used to discretize the continuous surfaces of the cavity, dividing the cavity region into several independent surface units. Each surface unit can be considered as the basic computational unit for local differential geometric analysis. For free-form cavities, an adaptive mesh refinement strategy is prioritized, increasing the mesh density in regions with drastic curvature changes to ensure accurate capture of surface geometric features. For each surface unit, a local parametric coordinate system is established with its center point or centroid as the origin. This coordinate system is typically orthogonal, with two coordinate axes lying in the tangent plane of the surface and the third axis along the direction of the surface normal vector. The establishment of the local parametric coordinate system provides a unified mathematical framework for subsequent surface differential geometric calculations.

[0031] Under the established local parametric coordinate system, the parametric equation expression of the surface element is extracted, assuming that the surface element can be expressed using parametric equations. The surface is represented as follows, where u and v are parametric coordinates. The first fundamental form coefficients of the surface are calculated, including three components: E, F, and G. These coefficients reflect the metric properties of the surface, where E represents the metric coefficient along the u direction, G represents the metric coefficient along the v direction, and F represents the cross metric coefficient along the u and v directions. These coefficients are obtained by taking partial derivatives of the surface parametric equations and calculating the inner product. Simultaneously, the second fundamental form coefficients L, M, and N are calculated, which describe the curvature of the surface. These coefficients are obtained by calculating the scalar product of the second-order partial derivatives of the surface and the unit normal vector. For discretized surface elements, a numerical differentiation method is used to approximate the partial derivatives, and a difference scheme is constructed using the positional information of adjacent mesh nodes.

[0032] Using the obtained first and second fundamental form coefficients, a shape operator matrix for the surface is constructed. The eigenvalues ​​of this matrix are the principal curvatures of the surface at that point, denoted as . and Principal curvature represents the maximum and minimum curvature values ​​of a surface at a given point along the principal directions; its physical meaning is the extreme value of the surface's curvature. The principal curvature is obtained by solving the characteristic equation, whose coefficients are composed of combinations of basic form coefficients. For numerical stability considerations, when the determinant of the first basic form coefficients approaches zero, it indicates that the surface is approaching a degenerate state in that region, requiring special treatment or mesh re-division. The sign of the principal curvature characterizes the convexity or concavity of the surface; positive values ​​indicate a convex surface, and negative values ​​indicate a concave surface.

[0033] Based on the calculated principal curvature and Then, the Gaussian curvature K of each surface element is calculated. Gaussian curvature is defined as the product of the two principal curvatures, i.e. Gaussian curvature is an intrinsic property of a surface and does not change with the surface's curvature in space. A positive Gaussian curvature indicates that the surface is elliptical, a negative Gaussian curvature indicates that the surface is saddle-shaped, and zero Gaussian curvature indicates that the surface is flat in at least one direction. The calculated principal curvatures... , The Gaussian curvature K is used as the curvature feature attribute of the surface element, and it is stored in the surface element data structure along with the geometric position, normal vector, area, and other information of the surface element. These curvature feature attributes constitute the basic data for subsequent surface feature boundary recognition.

[0034] After calculating the curvature characteristics of all surface elements, the common boundaries between adjacent surface elements within the cavity region are traversed. A common boundary is an edge segment shared by two adjacent surface elements, typically stored in the mesh data structure as edge objects or shared node pairs. For each common boundary, the principal curvature values ​​of the surface elements on both sides of the boundary are extracted. Since each surface element has two principal curvatures, appropriate principal curvatures need to be explicitly selected for comparison. Typically, the maximum principal curvature or the principal curvature along the boundary tangential direction is chosen as the comparison object. The gradient of the principal curvature variation of the surface elements on both sides of the common boundary is calculated. This gradient can be represented by the absolute value of the difference between the two principal curvatures or by a normalized relative rate of change. The magnitude of the principal curvature variation gradient directly reflects the degree of geometric continuity of the surface at that boundary.

[0035] The calculated gradient of principal curvature change is numerically compared with a pre-set curvature continuity threshold. The curvature continuity threshold is a criterion determined based on actual engineering requirements and model accuracy requirements, typically ranging from 5% to 20% of the curvature unit. When the gradient of principal curvature change exceeds this threshold, it indicates a significant curvature jump at the common boundary, breaking the geometric continuity of the surface; this common boundary is then identified as a surface feature boundary. The identification results of surface feature boundaries include the geometric location of the boundary, the degree of curvature difference on both sides of the boundary, and the boundary type identifier. These surface feature boundaries correspond to key geometric feature boundaries in cavity design, such as the intersection of fillets and planes, and transition zones between different curvature regions. Through the above-described differential geometric analysis and feature boundary identification process, accurate quantitative characterization of the geometric features of complex cavity surfaces is achieved, providing accurate surface geometric information for subsequent process planning.

[0036] In one optional implementation, based on the local curvature distribution characteristics of the cavity machining zone and material removal constraints, a multi-objective optimization model for tool trajectory planning is constructed, including: For each machining zone in the cavity machining zone, the curvature extrema and curvature change gradient within the machining zone are statistically analyzed to form a local curvature distribution feature characterizing the geometric complexity of the machining zone; Based on the local curvature distribution characteristics, the tool contact area morphology characteristics of the machining zone are determined, and the tool path length objective function and the machining surface quality uniformity objective function are established according to the tool contact area morphology characteristics. Obtain the material property parameters and tool geometry parameters in the machining process constraints, construct the material removal volume constraint per unit time based on the material property parameters and tool geometry parameters, and use the material removal volume constraint per unit time as the material removal constraint; Based on the material removal constraints, the tool interference avoidance space boundary and residual height control conditions are determined, and the tool interference avoidance space boundary and residual height control conditions are used as constraint conditions. By integrating the objective function of tool path length, the objective function of surface quality uniformity, and the constraints, a multi-objective optimization model with tool trajectory parameters as decision variables is constructed.

[0037] After obtaining the cavity machining partition results, a statistical analysis of the local curvature distribution characteristics is performed for each machining partition. All surface sampling points within the machining partition are traversed, and the principal curvature values ​​at each sampling point are extracted. The maximum and minimum curvature values ​​within the partition are determined as curvature extrema by comparison. Simultaneously, the curvature difference between adjacent sampling points is calculated, and the curvature change within a unit arc length is statistically analyzed to obtain the curvature gradient. The curvature gradient reflects the steepness of the surface shape; a larger value indicates a more drastic change in surface shape. The curvature extrema and curvature gradient are used together as characterization parameters of the local curvature distribution features to quantify the geometric complexity of the machining partition. When the curvature extrema are large or the curvature gradient is high, it indicates that the geometry of the partition is complex, placing higher demands on toolpath planning.

[0038] Based on the acquired local curvature distribution characteristics, the contact state between the tool and the workpiece surface is analyzed. When the surface curvature is large, the contact area between the tool and the workpiece exhibits a narrow elliptical shape; when the surface curvature is small, the contact area tends to be circular. By using the geometric relationship between curvature values ​​and tool radius, the major and minor axis dimensions of the contact area are calculated, determining the morphological characteristics of the tool contact area. Based on the morphological characteristics of the contact area, a toolpath length objective function is established. This function uses the total distance the tool travels within the machining zone as the optimization objective; a shorter travel distance indicates higher machining efficiency. When establishing the objective function for the uniformity of the machined surface quality, the residual height variance is introduced as an evaluation index. The residual height represents the height of material not removed between two adjacent tool passes. Specifically, the residual height is calculated as follows: at the center position between two adjacent toolpaths, a normal section perpendicular to the theoretically designed surface is established. Within this normal section, the tool sweep envelopes of the two adjacent toolpaths are extracted. The vertical distance between the lowest point of the two tool sweep envelopes and the theoretically designed surface is calculated; this vertical distance is the residual height at that position. For cylindrical or ball end mills, when the tool spacing is a e When the residual height h is reached, the formula for calculation is: , where R is the tool radius.

[0039] For different tool geometries, the residual height is obtained by solving the distance between the tool envelope and the theoretically designed surface, based on the geometric relationship between the tool profile equation and the tool path spacing. The statistical calculation method for the variance of the residual height is as follows: The entire cavity area is divided into several sampling points, each located at the center of an adjacent tool path. The residual height value of each sampling point is calculated in the above manner and denoted as h1, h2, ..., h n , where n is the total number of sampling points. Calculate the average residual height of all sampling points. Then calculate the variance of the residual height. The objective function for achieving uniformity of the machined surface quality is expressed as: By controlling the tool spacing to ensure consistent residual height at all points, even if the residual height variance σ 2 Minimizing this ensures the uniformity of the machined surface quality. The objective function optimizes by minimizing the variance of the residual height. A multi-objective optimization algorithm is used to adjust the row spacing parameter in the toolpath planning scheme, making the residual height in different areas more consistent, thereby achieving optimization of the uniformity of the machined surface quality.

[0040] The material property parameters involved in the machining process constraints are obtained, including the hardness, tensile strength, and specific energy of the material being machined. Tool geometry parameters include tool diameter, helix angle, rake angle, and clearance angle. Based on the material property parameters and tool geometry parameters, combined with the cutting speed and feed rate, the volume of material removed by the tool per unit time is calculated. The material removal volume is equal to the product of the depth of cut, the width of cut, and the feed rate. To ensure machining stability and avoid tool overload, the material removal volume per unit time must be controlled within a specific range. The upper limit of this range is determined by tool strength and machine tool power, while the lower limit is determined by machining efficiency requirements. The feasible range of material removal volume per unit time is expressed as an inequality constraint and incorporated into the optimization model as a material removal constraint.

[0041] The tool interference avoidance space boundary is determined based on material removal constraints. When the tool moves within the cavity, collisions between the non-cutting portion of the tool and the machined or unmachined surfaces must be avoided. The allowable spatial range of the tool center trajectory is determined by calculating the minimum clearance between the tool's geometric envelope and the cavity wall. The boundary of this spatial range is the tool interference avoidance space boundary, and the tool center must be located within this boundary during tool trajectory planning. Residual height control is achieved by limiting the maximum spacing between adjacent tool paths. The maximum allowable row spacing is calculated based on the tool radius and the target residual height. When the tool moves along a parallel path, the spacing between adjacent paths must not exceed this maximum row spacing; otherwise, the residual height will exceed the allowable range, affecting the quality of the machined surface. The tool interference avoidance space boundary is expressed as a spatial constraint on the tool center coordinates, and the residual height control condition is expressed as an upper bound constraint on the spacing between adjacent paths. Together, these constitute the geometric and technological constraints of the optimization model.

[0042] The established objective functions for toolpath length and surface quality uniformity are integrated to form a set of objective functions for a multi-objective optimization model. These two objective functions have a certain conflict: shortening the toolpath length increases the spacing between adjacent paths, thus reducing surface quality uniformity. Tool trajectory parameters are selected as decision variables, including the starting point coordinates of the toolpath, the path direction angle, the spacing between adjacent paths, and the tool feed rate. The tool interference avoidance space boundary, residual height control conditions, and material removal constraints are collectively used as a set of constraints. These constraints limit the feasible domain of the decision variables, ensuring that the generated toolpath meets both geometric feasibility and process requirements. The constructed multi-objective optimization model aims to find the Pareto optimal solution set, achieving a relatively good state for both toolpath length and surface quality uniformity. This optimization model provides the mathematical foundation for subsequent toolpath generation; solving this model yields a toolpath scheme that balances machining efficiency and surface quality. The dimensions of all parameters in the optimization model are kept consistent: path length is expressed in millimeters, and residual height variance is expressed in square millimeters, ensuring the comparability of the objective functions. The degree of relaxation of constraints can be adjusted according to the actual machining accuracy requirements. In high-precision machining scenarios, the residual height constraint is tightened, while the material removal constraint is appropriately relaxed in the roughing stage, so as to achieve flexible process adaptability.

[0043] In one optional implementation, the multi-objective optimization model is solved through adaptive iteration. During the iteration process, the path generation strategy between adjacent machining zones is dynamically adjusted based on the tool trajectory simulation results of the previous iteration. This generates a tool trajectory planning scheme that includes the entry point position, cutting direction, and line spacing parameters, including: Initialize the decision variables of the multi-objective optimization model, iteratively solve the multi-objective optimization model, and obtain the initial tool trajectory candidate scheme; The initial tool trajectory candidate scheme is simulated to calculate the tool's motion trajectory in each machining zone and obtain the tool trajectory simulation results of the previous iteration. Extract the tool posture continuity deviation from the tool trajectory simulation results, compare the tool posture continuity deviation with a preset continuity threshold, and when the tool posture continuity deviation exceeds the preset continuity threshold, determine that there is a need for connection path adjustment between adjacent machining zones; For adjacent machining partitions that have the required connection path adjustment, the residual error of the machining surface at the corresponding boundary position in the tool trajectory simulation result is extracted, the distribution characteristics of the residual error of the machining surface are analyzed, the connection path generation strategy between the adjacent machining partitions is dynamically adjusted according to the distribution characteristics, and the row spacing parameter of the adjacent machining partitions is adjusted based on the tool posture continuity deviation. The adjusted connection path generation strategy and the row spacing parameter are updated to the decision variables. Iterative solution and simulation steps are repeated until the tool posture continuity deviation and the residual error of the machining surface meet the preset convergence conditions, and the tool trajectory planning scheme is output.

[0044] After the multi-objective optimization model is established, it needs to be solved through an adaptive iterative mechanism to obtain the optimal toolpath planning scheme. The core of this process is to use the simulation feedback information from the previous iteration to guide the parameter adjustment in subsequent iterations, especially to dynamically optimize the connection path generation strategy between adjacent machining zones.

[0045] In the initial stage of iterative solution, the decision variables of the multi-objective optimization model are initialized. These variables mainly include the spatial coordinates of the tool entry point in each machining zone, the unit vector of the tool cutting direction, the line spacing between adjacent tool paths, and the type identifier of the connecting paths. Initialization can employ heuristic methods based on surface geometry features. For example, the tool entry point can be set in a region with minimal curvature change in each zone, the cutting direction can be along the principal curvature direction, and the line spacing can be initialized according to machining accuracy requirements. After initialization, a genetic algorithm or particle swarm optimization algorithm is used to perform the first round of iterative solution for the multi-objective optimization model. Under the premise of satisfying various constraints, the decision space is searched to obtain several initial tool trajectory candidate schemes. These candidate schemes are distributed on the Pareto front, representing different trade-offs between machining efficiency and machining quality.

[0046] After obtaining initial candidate toolpath schemes, detailed tool motion simulations are performed for each scheme. The simulation process requires constructing a virtual machining environment that includes a tool geometry model, a workpiece surface model, and a machine tool kinematics model. Based on the entry point position, cutting direction, and path spacing parameters specified in the candidate schemes, the continuous motion trajectory of the tool within each machining zone is calculated. The simulation calculations need to consider the contact state between the tool and the surface, including the spatial direction of the tool axis vector, the surface coordinates of the tool contact point, and the changes in the tool attitude angle. Special attention is paid to the tool motion state at the boundary between adjacent machining zones, recording the attitude change process of the tool transitioning from one zone to another. The simulation results are stored as a dataset, containing the position coordinates, attitude angles, cutting speed, and deviations from the theoretical surface for each discrete point on the toolpath.

[0047] After obtaining the tool trajectory simulation results from the previous iteration, key quality evaluation indicators need to be extracted for subsequent strategy adjustments. Tool attitude continuity deviation is one of the important evaluation indicators, and its calculation method is to extract the change in tool attitude angle at the boundary of adjacent machining zones. Specifically, when the tool moves from the end point of one machining zone to the start point of an adjacent zone, the change angle of the tool axis vector direction and the change of the tool's rotation angle around its own axis are calculated. The maximum value of these angle changes is used as the quantitative indicator of tool attitude continuity deviation. A preset continuity threshold of 3 to 5 degrees is set. This threshold range is determined based on the dynamic response characteristics of the five-axis machining center and the smoothness requirements of surface machining. The calculated tool attitude continuity deviation is compared with the preset continuity threshold one by one. When the deviation at a certain point exceeds the threshold, it is determined that there is a need to adjust the connection path between adjacent machining zones at that location, and the connection strategy at that point needs to be optimized in the next iteration.

[0048] For adjacent machining zones identified as requiring adjustment of the connection path, more detailed machining quality information is extracted. The corresponding boundary positions are located from the tool trajectory simulation results, and residual error data of the machining surface in that area is extracted. Residual error refers to the deviation of the normal distance between the tool sweep envelope and the theoretically designed surface. Significant residual errors occur in the boundary region due to improper tool trajectory connection. Spatial distribution characteristic analysis is performed on the extracted residual error data, calculating statistical features including mean, standard deviation, maximum value, and spatial gradient of the error. The spatial pattern of the error distribution is analyzed to determine if there are systematic overcutting or undercutting phenomena. Based on the analysis results, the connection path generation strategy between adjacent machining zones is dynamically adjusted. Adjustments include modifying the geometry of the connection path, adjusting the length of the transition segment, and optimizing the tool approach and retraction methods. For example, when overcutting occurs at the boundary, straight-line connections can be changed to arc transition connections, and the transition segment length can be increased to achieve smoother tool posture changes; when undercutting occurs, compensation paths can be added at the boundary or the boundary positions of adjacent zones can be adjusted.

[0049] The row spacing parameters of adjacent machining zones are adjusted based on the specific numerical value of the tool attitude continuity deviation. The row spacing parameters directly affect the toolpath density and the residual height of the machined surface. When the tool attitude continuity deviation is large, it indicates that the toolpath connection between adjacent zones is not smooth enough. The connection quality can be improved by reducing the row spacing to increase the path density. The adjustment strategy uses a proportional adjustment method. A row spacing adjustment coefficient is calculated based on the extent to which the attitude continuity deviation exceeds a threshold. This coefficient is typically between 0.8 and 0.95. The original row spacing is multiplied by this coefficient to obtain the new row spacing value. Row spacing adjustment must also consider the impact on machining efficiency to avoid excessively reducing the row spacing, which could lead to a significant increase in machining time.

[0050] The adjusted path generation strategy and updated line spacing parameters are reassigned to the decision variables of the multi-objective optimization model, forming the input conditions for a new round of iteration. The multi-objective optimization solution process is executed again, searching for a better tool trajectory scheme in the updated decision space. The tool motion simulation steps are repeated for the newly generated tool trajectory candidate schemes to calculate the new tool trajectory simulation results and extract the tool attitude continuity deviation and machining surface residual error. A convergence evaluation mechanism is established, setting the preset convergence conditions as follows: the tool attitude continuity deviation is less than a preset continuity threshold and the maximum value of the machining surface residual error is less than a given residual error limit, while the rate of change of the objective function value between two consecutive iterations is less than 0.5%. After each iteration, it is determined whether the convergence conditions are met. If not, the next round of iteration adjustment continues; if met, the iteration process is terminated.

[0051] Once the iterative process converges, a set of solutions with optimal overall performance is selected from the final Pareto solution set as the output toolpath planning scheme. This scheme fully includes the three-dimensional coordinates of the entry point of each machining zone, the unit vector components of the tool cutting direction, the optimized line spacing parameter values, and the geometric description of the connection path determined through multiple rounds of adjustments.

[0052] In one optional implementation, tool motion simulation is performed on the initial tool trajectory candidate scheme to calculate the tool's motion trajectory within each machining zone, obtaining the tool trajectory simulation results of the previous iteration, including: Extract the tool path node coordinate sequence of each machining zone in the initial tool path candidate scheme, and construct the tool center point motion trajectory based on the tool path node coordinate sequence; Based on the motion trajectory of the tool center point and combined with the tool's geometric parameters, the tool attitude vector at each path node position is calculated. Based on the tool posture vector, calculate the tool posture change between adjacent path nodes, and record the value of the tool posture change at the boundary position of adjacent machining zones as the tool posture continuity deviation. Based on the tool center point motion trajectory and the tool attitude vector, a tool sweep body model is established. Boolean operation is performed on the tool sweep body model to calculate the cavity surface morphology after the tool sweep body model is removed. The deviation value between the cavity surface morphology and the theoretically designed surface is extracted, and the deviation value is used as the residual error of the machining surface. The tool posture continuity deviation and the residual error of the machined surface are combined to form the tool trajectory simulation results of the previous iteration.

[0053] When performing motion simulation on the initial candidate toolpath schemes, node coordinate information is extracted from the toolpath data of each machining zone. The toolpath of each machining zone consists of a series of discrete spatial coordinate points, arranged in sequence according to the machining order. The extraction process requires parsing the path file format and extracting the three-dimensional coordinates of each node. Read and store sequentially, where the subscript i represents the node number. For multi-partition machining of complex cavities, it is necessary to extract the coordinate sequence of each partition and establish a record of the topological relationship between the partitions.

[0054] After obtaining the toolpath node coordinate sequence, a continuous motion trajectory of the tool center point is constructed using interpolation methods. For trajectory segments between adjacent nodes, spline interpolation is used to generate smooth spatial curves. Specifically, a cubic spline interpolation algorithm can be selected. This algorithm ensures that the curve passes through all nodes while guaranteeing that adjacent curve segments have continuous first and second derivatives at the connection points. During interpolation, a reasonable interpolation density needs to be set, typically by adaptively adjusting the number of interpolation points based on the distance between adjacent nodes to ensure that the curve accurately reflects the actual tool motion path. For special geometric shapes such as straight lines and circular arcs, the coordinates of intermediate points can be directly calculated using the corresponding analytical expressions, avoiding unnecessary interpolation calculations.

[0055] The calculation of the tool attitude vector requires combining the tool's geometric parameters and the trajectory information of the tool's center point. Tool attitude is typically represented by a tool axis direction vector, which describes the tool's orientation in space. For ball end mills, the attitude vector points from the tool's center point to the tool axis; for flat end mills, the attitude vector is perpendicular to the tool's bottom surface. When calculating the tool attitude at each path node, the theoretical machining attitude of the tool is first determined based on the normal vector of the cavity surface at that point. Then, the attitude is adjusted considering the interference between the tool and the workpiece. The adjustment process involves detecting the relative positional relationship between the tool's envelope surface and the cavity surface. When potential interference is detected, the tool attitude angle is fine-tuned within permissible limits to ensure effective cutting without collision with the workpiece. The tool attitude vector is usually stored as a unit vector for easy calculation of subsequent attitude changes.

[0056] Tool attitude continuity assessment is achieved by calculating the attitude changes between adjacent path nodes. For two adjacent nodes, their tool attitude vectors are obtained respectively. and The angle between two vectors is calculated as the attitude change. The angle is calculated using vector dot product, and the angle value is obtained through the inverse cosine function. At the boundary of machining zones, special attention needs to be paid to the continuity of the tool attitude. Due to differences in path planning strategies between different zones, abrupt changes in tool attitude can easily occur at the zone boundaries. When extracting the attitude change of zone boundary nodes, this change is compared with a preset attitude continuity threshold. Changes exceeding the threshold are recorded as tool attitude continuity deviations. This deviation reflects the smoothness of the tool path during zone transitions; excessive deviation can lead to significant feed rate fluctuations or surface quality changes during machining.

[0057] Establishing a tool sweep volume model is a crucial step in accurately predicting the surface morphology of the machined surface. The sweep volume represents the spatial volume swept by the tool along its motion trajectory, and the boundary of this volume is the set of all surface points that the tool can reach. During modeling, the tool geometry is swept along the motion trajectory of the tool's center point, while the spatial orientation of the tool is adjusted according to the tool's attitude vector at each position. For ball end mills, the cross-section of the sweep volume is circular, and the longitudinal profile is determined by both the motion trajectory and attitude changes; for flat end mills, the sweeping effects on the bottom and side surfaces need to be considered simultaneously. The accurate construction of the sweep volume employs a discretization method, decomposing the continuous tool motion into multiple small motion steps. At each step position, the instantaneous geometry of the tool is generated, and then Boolean union operations are used to merge all instantaneous shapes into a complete sweep volume.

[0058] Boolean operations are used to calculate the material removal effect of the tool sweep body on the cavity blank. The initial geometric model of the cavity blank is taken as the subtracted object, and the tool sweep body model is taken as the subtracted object; a Boolean difference operation is performed. The result is the surface morphology model of the machined cavity. The Boolean operation process needs to handle complex geometric intersection problems, especially when there are multiple intersection lines between the sweep body and the blank surface, requiring accurate determination of the topological relationship of the intersection lines and the material removal direction. After the operation is completed, a triangular mesh representation of the cavity surface is extracted; this mesh contains all the geometric details of the machined surface.

[0059] Residual errors on the machined surface are obtained by comparing the actual machined surface with the theoretically designed surface. Point-to-point distance measurements are performed between the cavity surface morphology obtained from Boolean operations and the theoretically designed surface in the CAD model. During measurement, measurement points are uniformly sampled on the theoretically designed surface, and the shortest distance from each measurement point to the actual machined surface is calculated. The distance calculation uses a point-to-grid projection algorithm to find the nearest projection position of the measurement point on the machined surface grid, and the Euclidean distance between the two points is calculated as the residual error value at that position. Positive deviation indicates overcutting, and negative deviation indicates undercutting. The deviation values ​​of all measurement points are statistically analyzed, and the maximum deviation, average deviation, and deviation distribution characteristics are extracted. These values ​​comprehensively reflect the machining accuracy level of the current toolpath scheme.

[0060] The simulation results are summarized by integrating tool attitude continuity deviation and residual error on the machined surface into comprehensive evaluation data. Attitude continuity deviation is recorded as a list of angle changes at the boundaries of each zone, along with the specific coordinates of any abnormal deviations. Residual error on the machined surface is presented as statistical data and error cloud maps. The error cloud maps visually display the distribution of machining quality in different areas of the cavity surface using color mapping. These data, along with auxiliary information such as the time series of the tool motion trajectory and feed rate variation curves, are stored to form a complete simulation result of the previous iteration's tool trajectory, providing detailed analytical basis for subsequent iterative optimization.

[0061] In one optional implementation, the tool path planning scheme is converted into CNC machining instructions and output to the machining equipment. Simultaneously, actual cutting force feedback information during the machining process is collected. The material removal constraint parameters in the multi-objective optimization model are then corrected using this actual cutting force feedback information, including: The tool path planning scheme is analyzed, including the entry point position, cutting direction, and line spacing parameters. Numerical control machining instructions are generated based on the entry point position, cutting direction, and line spacing parameters. The CNC machining instructions are output to the machining equipment to control the machining equipment to perform cavity machining operations; During the machining operation, the cutting force data generated when the tool comes into contact with the workpiece is collected in real time by a force sensor. The cutting force data includes the cutting force amplitude and the cutting force fluctuation characteristics. A time-series analysis is performed on the cutting force data to extract the average value of the cutting force amplitude and the peak value of the cutting force fluctuation characteristics. The average value and the peak value are used as actual cutting force feedback information. The preset theoretical cutting force is obtained from the multi-objective optimization model, and the actual cutting force feedback information is compared with the theoretical cutting force to calculate the deviation between the actual cutting force feedback information and the theoretical cutting force. Based on the deviation, the material removal constraint parameters in the multi-objective optimization model are corrected.

[0062] After the toolpath planning scheme is generated, it needs to be converted into a CNC instruction format that the machining equipment can recognize. The toolpath planning scheme is analyzed to extract the spatial coordinates of the entry point, the cutting direction vector, and the line spacing values ​​between adjacent toolpaths. The entry point position is usually represented in three-dimensional coordinates, recording the accurate position of the tool's initial contact with the workpiece; the cutting direction describes the tool's movement on the cavity surface through direction angles or vectors; the line spacing parameter defines the interval between two adjacent toolpaths, directly affecting the machining quality of the cavity surface.

[0063] Based on the extracted parameter information, and in accordance with the ISO 6983 standard or the instruction format requirements of a specific CNC system, a CNC machining instruction sequence containing G-code, M-code, and coordinate data is generated. The instruction sequence includes tool rapid positioning instructions, feed rate setting instructions, spindle speed control instructions, and toolpath interpolation instructions. The tool entry point position is converted to absolute or incremental coordinates, and the cutting direction is achieved through linear or circular interpolation instructions. The line spacing parameter is reflected in the offset of the starting coordinates of adjacent path segments. The generated CNC machining instructions are transmitted to the CNC system of the machining equipment via a standard communication interface. After parsing the instructions, the CNC system drives the servo motors of each axis, controlling the tool to mill the cavity according to the planned trajectory.

[0064] While the machining equipment performs cavity machining operations, a multi-dimensional force sensor is installed between the machine tool table and the workpiece fixture to monitor the changes in cutting force in real time during the machining process. The force sensor adopts the piezoelectric or strain gauge measurement principle and can simultaneously collect three-dimensional force data of the tool in the feed direction, depth of cut direction, and axial direction. The sensor continuously records the cutting force signal at a high sampling frequency, which is usually set between 1000Hz and 5000Hz to ensure the capture of transient force changes during the cutting process.

[0065] The acquired raw cutting force signal contains effective cutting force components and noise interference components. The raw signal undergoes digital filtering, employing a low-pass filter to remove high-frequency noise and retain the force signal components reflecting the actual cutting state. Two key features are extracted from the filtered cutting force signal: the cutting force amplitude reflects the overall load level borne by the tool, and the cutting force fluctuation characteristics reflect the stability of the cutting process. The cutting force amplitude is obtained by calculating the composite vector of the three-dimensional force components, and the cutting force fluctuation characteristics are obtained by analyzing the time-domain rate of change or frequency-domain energy distribution of the force signal.

[0066] Time-series analysis is performed on the processed cutting force data. Within a set time window, the arithmetic mean of the cutting force amplitude sequence is calculated. This mean represents the average cutting load borne by the tool during that time period. Simultaneously, peak detection is performed on the cutting force fluctuation characteristic sequence to identify the maximum value of the fluctuation amplitude. This peak reflects the most unfavorable load condition encountered during the cutting process. The calculated average cutting force amplitude and the peak value of the cutting force fluctuation characteristic sequence together constitute the actual cutting force feedback information, which truly reflects the cutting force state under the current machining conditions.

[0067] In the multi-objective optimization model establishment stage, theoretical cutting force predictions are obtained through empirical formulas or finite element simulation calculations based on tool geometry parameters, material mechanical properties, and cutting parameters. The theoretical cutting force is based on ideal machining conditions and does not consider the influence of actual factors such as tool wear, material hardness fluctuations, and clamping stiffness changes. Pre-set theoretical cutting force values ​​are extracted from the constraint condition module of the optimization model and compared with real-time acquired actual cutting force feedback information.

[0068] The deviation between the actual cutting force feedback information and the theoretical cutting force is calculated. This deviation includes both the absolute numerical difference and the relative percentage error. When the average amplitude of the actual cutting force is significantly higher than the theoretical cutting force, it indicates that the current combination of cutting parameters is causing tool overload; when the peak value of the cutting force fluctuation exceeds the theoretical prediction range, it indicates the presence of unstable factors in the cutting process. By analyzing the magnitude and sign of the deviation, the rationality of the material removal constraint parameters set in the multi-objective optimization model is determined.

[0069] Based on the calculated deviation, the material removal constraint parameters in the multi-objective optimization model are corrected. These parameters mainly include the upper limit of material removal rate per unit time, the upper limit of single-cut layer thickness, and the allowable cutting force threshold. When the deviation is positive and exceeds a preset threshold, the upper limit of the material removal rate constraint or the allowable cutting force threshold is reduced, allowing the optimization model to generate more conservative cutting parameter combinations in subsequent iterations. When the deviation is negative, the constraints are appropriately relaxed to improve machining efficiency. The corrected constraint parameters are then rewritten into the constraint equations of the optimization model, providing a more accurate optimization basis for the next batch of cavity machining or subsequent areas of the same cavity. Through this closed-loop feedback correction mechanism, the matching accuracy between the tool trajectory planning scheme and actual machining conditions is gradually improved, achieving adaptive optimization control of the cavity machining process. The correction process continues until the deviation converges to an acceptable range, ensuring that machining quality and efficiency meet the expected goals.

[0070] A second aspect of the present invention provides an intelligent planning and optimization system for injection mold processing paths, comprising: The cavity partitioning unit is used to obtain the three-dimensional geometric model of the cavity to be processed and the processing constraints. Based on the surface features of the three-dimensional geometric model, the geometric topology information of the cavity region is extracted, and the geometric topology information is partitioned according to the processing constraints to obtain the cavity processing partition with processing priority identifier. The trajectory optimization unit is used to construct a multi-objective optimization model for tool trajectory planning based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints. The iterative solution unit is used to solve the multi-objective optimization model through adaptive iteration. During the iteration process, it dynamically adjusts the connection path generation strategy between adjacent machining zones based on the tool trajectory simulation results of the previous iteration, and generates a tool trajectory planning scheme that includes the entry point position, cutting direction and line spacing parameters. The instruction output unit is used to convert the tool trajectory planning scheme into CNC machining instructions and output them to the machining equipment. At the same time, it collects the actual cutting force feedback information during the machining process and uses the actual cutting force feedback information to correct the material removal constraint parameters in the multi-objective optimization model.

[0071] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0072] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0073] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0074] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. An intelligent planning and optimization method for injection mold processing paths, characterized in that, include: Obtain the three-dimensional geometric model of the cavity to be processed and the processing constraints. Extract the geometric topology information of the cavity region based on the surface features of the three-dimensional geometric model. Then, partition the geometric topology information according to the processing constraints to obtain the cavity processing partition with processing priority identifier. Based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints, a multi-objective optimization model for tool trajectory planning is constructed. The multi-objective optimization model is solved by adaptive iteration. During the iteration process, the connection path generation strategy between adjacent machining zones is dynamically adjusted based on the tool trajectory simulation results of the previous iteration, and a tool trajectory planning scheme including the entry point position, cutting direction and line spacing parameters is generated. The tool path planning scheme is converted into CNC machining instructions and output to the machining equipment. At the same time, the actual cutting force feedback information during the machining process is collected, and the material removal constraint parameters in the multi-objective optimization model are corrected using the actual cutting force feedback information.

2. The method according to claim 1, characterized in that, Based on the surface features of the three-dimensional geometric model, the geometric topology information of the cavity region is extracted, and the geometric topology information is partitioned according to the processing constraints to obtain cavity processing partitions with processing priority identifiers, including: The three-dimensional geometric model is subjected to surface differential geometry analysis to calculate the principal curvature and Gaussian curvature of each surface unit in the cavity region. The surface feature boundary is identified based on the curvature continuity condition, and geometric topological information containing surface type identifier and boundary topological relationship is extracted. Based on the surface type identifier in the geometric topology information, and combined with the tool accessibility condition and machining direction restriction in the machining process constraints, the cavity region is spatially decomposed, and adjacent surface units with the same machining direction feasibility are merged into the same machining partition. Based on the spatial relationship between the processing partitions and the boundary topology, the processing dependency relationship between each processing partition is determined, and a processing priority identifier is assigned to each processing partition according to the processing dependency relationship, thus obtaining cavity processing partitions with processing priority identifiers.

3. The method according to claim 2, characterized in that, The three-dimensional geometric model is subjected to surface differential geometry analysis to calculate the principal curvature and Gaussian curvature of each surface element in the cavity region, and the surface feature boundaries are identified based on the curvature continuity condition, including: The cavity region of the three-dimensional geometric model is discretized into a surface mesh, dividing the cavity region into multiple surface units, and a local parametric coordinate system is established for each surface unit. In the local parameter coordinate system, the first basic form coefficient and the second basic form coefficient of each surface element are calculated, and the principal curvature of the surface element is solved using the first basic form coefficient and the second basic form coefficient. The Gaussian curvature of each surface element is calculated based on the principal curvature, and the principal curvature and the Gaussian curvature are stored as curvature feature attributes of the surface element. Traverse the common boundary between adjacent surface elements, extract the principal curvature of the surface elements on both sides of the common boundary, and calculate the gradient of the principal curvature change of the surface elements on both sides of the common boundary. The principal curvature change gradient is compared with a preset curvature continuity threshold. When the principal curvature change gradient exceeds the curvature continuity threshold, the common boundary is determined to be a surface feature boundary.

4. The method according to claim 1, characterized in that, Based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints, a multi-objective optimization model for tool trajectory planning is constructed, including: For each machining zone in the cavity machining zone, the curvature extrema and curvature change gradient within the machining zone are statistically analyzed to form a local curvature distribution feature characterizing the geometric complexity of the machining zone; Based on the local curvature distribution characteristics, the tool contact area morphology characteristics of the machining zone are determined, and the tool path length objective function and the machining surface quality uniformity objective function are established according to the tool contact area morphology characteristics. Obtain the material property parameters and tool geometry parameters in the machining process constraints, construct the material removal volume constraint per unit time based on the material property parameters and tool geometry parameters, and use the material removal volume constraint per unit time as the material removal constraint; Based on the material removal constraints, the tool interference avoidance space boundary and residual height control conditions are determined, and the tool interference avoidance space boundary and residual height control conditions are used as constraint conditions. By integrating the objective function of tool path length, the objective function of surface quality uniformity, and the constraints, a multi-objective optimization model with tool trajectory parameters as decision variables is constructed.

5. The method according to claim 1, characterized in that, The multi-objective optimization model is solved through adaptive iteration. During the iteration process, the path generation strategy between adjacent machining zones is dynamically adjusted based on the tool trajectory simulation results of the previous iteration. This generates a tool trajectory planning scheme that includes parameters such as the entry point position, cutting direction, and line spacing. Initialize the decision variables of the multi-objective optimization model, iteratively solve the multi-objective optimization model to obtain an initial tool trajectory candidate scheme; perform tool motion simulation on the initial tool trajectory candidate scheme, calculate the motion trajectory of the tool in each machining zone, and obtain the tool trajectory simulation results of the previous iteration; Extract the tool posture continuity deviation from the tool trajectory simulation results, compare the tool posture continuity deviation with a preset continuity threshold, and when the tool posture continuity deviation exceeds the preset continuity threshold, determine that there is a need for connection path adjustment between adjacent machining zones; For adjacent machining partitions that have the required connection path adjustment, the residual error of the machining surface at the corresponding boundary position in the tool trajectory simulation result is extracted, the distribution characteristics of the residual error of the machining surface are analyzed, the connection path generation strategy between the adjacent machining partitions is dynamically adjusted according to the distribution characteristics, and the row spacing parameter of the adjacent machining partitions is adjusted based on the tool posture continuity deviation. The adjusted connection path generation strategy and the row spacing parameter are updated to the decision variables. Iterative solution and simulation steps are repeated until the tool posture continuity deviation and the residual error of the machining surface meet the preset convergence conditions, and the tool trajectory planning scheme is output.

6. The method according to claim 5, characterized in that, The initial tool trajectory candidate scheme is simulated to calculate the tool's motion trajectory in each machining zone. The simulation results of the previous iteration of the tool trajectory are obtained as follows: Extract the tool path node coordinate sequence of each machining zone in the initial tool path candidate scheme, and construct the tool center point motion trajectory based on the tool path node coordinate sequence; Based on the motion trajectory of the tool center point and combined with the tool's geometric parameters, the tool attitude vector at each path node position is calculated. Based on the tool posture vector, calculate the tool posture change between adjacent path nodes, and record the value of the tool posture change at the boundary position of adjacent machining zones as the tool posture continuity deviation. Based on the tool center point motion trajectory and the tool attitude vector, a tool sweep body model is established. Boolean operation is performed on the tool sweep body model to calculate the cavity surface morphology after the tool sweep body model is removed. The deviation value between the cavity surface morphology and the theoretically designed surface is extracted, and the deviation value is used as the residual error of the machining surface. The tool posture continuity deviation and the residual error of the machined surface are combined to form the tool trajectory simulation results of the previous iteration.

7. The method according to claim 1, characterized in that, The tool path planning scheme is converted into CNC machining instructions and output to the machining equipment. Simultaneously, actual cutting force feedback information during the machining process is collected. The material removal constraint parameters in the multi-objective optimization model are then corrected using this actual cutting force feedback information, including: The tool path planning scheme is analyzed, including the entry point position, cutting direction, and line spacing parameters. Numerical control (NC) machining instructions are generated based on the entry point position, cutting direction, and line spacing parameters. The NC machining instructions are then output to the machining equipment to control the machining equipment to perform cavity machining operations. During the machining operation, a force sensor collects cutting force data generated when the tool contacts the workpiece in real time. The cutting force data includes the cutting force amplitude and the cutting force fluctuation characteristics. The cutting force data is subjected to time series analysis to extract the average value of the cutting force amplitude and the peak value of the cutting force fluctuation characteristics. The average value and the peak value are used as actual cutting force feedback information. A preset theoretical cutting force is obtained from the multi-objective optimization model. The actual cutting force feedback information is compared with the theoretical cutting force, and the deviation between the actual cutting force feedback information and the theoretical cutting force is calculated. Based on the deviation, the material removal constraint parameters in the multi-objective optimization model are corrected.

8. An intelligent planning and optimization system for injection mold processing paths, used to implement the method described in any one of claims 1-7, characterized in that, include: The cavity partitioning unit is used to obtain the three-dimensional geometric model of the cavity to be processed and the processing constraints. Based on the surface features of the three-dimensional geometric model, the geometric topology information of the cavity region is extracted, and the geometric topology information is partitioned according to the processing constraints to obtain the cavity processing partition with processing priority identifier. The trajectory optimization unit is used to construct a multi-objective optimization model for tool trajectory planning based on the local curvature distribution characteristics of the cavity machining zone and the material removal constraints. The iterative solution unit is used to solve the multi-objective optimization model through adaptive iteration. During the iteration process, it dynamically adjusts the connection path generation strategy between adjacent machining zones based on the tool trajectory simulation results of the previous iteration, and generates a tool trajectory planning scheme that includes the entry point position, cutting direction and line spacing parameters. The instruction output unit is used to convert the tool trajectory planning scheme into CNC machining instructions and output them to the machining equipment. At the same time, it collects the actual cutting force feedback information during the machining process and uses the actual cutting force feedback information to correct the material removal constraint parameters in the multi-objective optimization model.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images