A visualization acceleration method for large material removal complex workpiece machining simulation

Abstract

This invention relates to a visualization acceleration method for simulating the machining of complex workpieces with large material removal volumes, belonging to the field of mechanical engineering. First, a workpiece modeling method based on uniform voxels is established. Second, the maximum tool feed step size in three-axis machining mode is derived, and discrete tool positions are obtained through interpolation. Third, a material removal method based on a three-dimensional distance field is established. Fourth, a workpiece surface triangular mesh visualization method is established. Fifth, a static-dynamic voxel separation method is established, and the previous methods are integrated to form the final proposed material removal process simulation acceleration method. Based on static-dynamic voxel separation, this invention achieves dual decoupling of static and dynamic voxels in time and space, effectively controlling the number of dynamic voxels in each formal cutting simulation segment, significantly reducing the visualization workload of the simulation process, and solving the problem of low visualization efficiency in traditional voxel-based machining process material removal simulation, thus achieving rapid simulation of the machining process.

Description

Technical Field

[0001] This invention belongs to the field of mechanical engineering and relates to a simulation method for the material removal process in machining, and in particular to a visualization acceleration method for simulating the machining of complex workpieces with large material removal volumes. Background Technology

[0002] Material removal simulation in CNC machining is a key technology in mechanical engineering. By simulating the cutting process of a tool on a workpiece, the rationality of the toolpath can be verified, the surface quality of the workpiece can be checked, and appropriate tools and cutting parameters can be selected. To date, various methods for simulating material removal in machining processes have been proposed. Among them, the voxel method uses uniformly sized voxels to model the workpiece. Besides its high accuracy, this method also utilizes the strict spatial isotropy of voxels to provide a consistent and complete mathematical representation of the physical properties of the entire workpiece. Therefore, voxel-based modeling has become a mainstream method in material removal simulation.

[0003] A significant drawback of the voxel method is that simulation accuracy and efficiency are directly related to the number of voxels. To ensure model accuracy, the computational inefficiency caused by the large number of voxels must be addressed. In fact, due to tool bounding box technology, the time spent on cutting calculations after each tool movement is very short; most of the time is spent on visually updating the surface triangular mesh of the voxels after cutting. This is because traditional methods such as the McLeod method extract the surface triangular mesh of the entire workpiece, making the extraction of most of the unchanged workpiece surface mesh a repetitive process. This problem is even more severe when simulating the machining of complex workpieces with large material removal volumes.

[0004] To address the above issues, scholars both domestically and internationally have conducted extensive research and proposed several solutions. For example, Chinese invention patent (application number CN202211442374.X) proposes a dynamic voxel partitioning method based on toolpath look-ahead. This method calculates intersecting voxels on the look-ahead toolpath that intersect with the tool, further partitions these voxels, and combines them with Tri-dexel models to achieve efficient geometric cutting simulation. Another Chinese invention patent (application number CN202510439658.0) proposes an efficient cutting simulation method based on an octree-accelerated Tri-dexel model. This method groups and stores Dexel rays according to the depth of the octree, and performs cutting through one-dimensional Boolean operations on Dexel segments between the workpiece and tool Tri-dexel models, effectively reducing the computational load and storage space required for machining simulation. While existing technologies can improve simulation speed, most rely on multi-level partitioned voxel models or Tri-dexel models. Multi-level voxel models (such as octrees) require voxel subdivision operations during each cut, leading to a significant increase in Boolean operation time. Furthermore, computational efficiency decreases as the number of small voxels increases. Tri-dexel-based models, on the other hand, introduce the anisotropy of Dexel rays, making it difficult to characterize workpiece physical properties as effectively as fully isotropic voxel models.

[0005] Therefore, it is necessary to propose a visualization-based acceleration method for machining simulation of complex workpieces composed of uniform voxels with large material removal volumes. Summary of the Invention

[0006] This invention addresses the shortcomings of traditional voxel-based material removal simulation methods in terms of visualization efficiency. In the field of machining material removal process simulation, it proposes a visualization acceleration method for simulating the machining of complex workpieces with large material removal volumes based on static-dynamic voxel separation. This invention can effectively reduce the number of voxels that need to be visualized in the simulation by separating static and dynamic voxels in the simulation process, thereby achieving a significant improvement in simulation efficiency.

[0007] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes is proposed. This method is based on the core concept of static-dynamic voxel separation and is combined with several fundamental material removal simulation methods. First, a workpiece modeling method based on uniform voxels is established. Second, the maximum tool feed step size in three-axis machining mode is derived, and discrete tool positions are obtained through interpolation. Third, a material removal method based on a three-dimensional distance field is established. Fourth, a workpiece surface triangular mesh visualization method is established. Fifth, a static-dynamic voxel separation method is established, and the previous methods are integrated to form the final proposed material removal process simulation acceleration method. The method includes the following steps: S1: Establish a workpiece modeling method based on uniform voxels to obtain discretized voxel models of classic cuboid and cylindrical blanks. Specifically: Model two classic stock geometries used in five-axis machining using voxels of uniform size. These stock geometries include cuboid and cylindrical stock geometries. For the cuboid stock, its length and voxel dimensions in the X, Y, and Z directions need to be input. Discretize and voxelize in three directions to obtain the voxel model of the cuboid blank. For the cylindrical blank, after inputting its diameter, axial length, central axis direction, and voxel dimensions, first establish a cuboid bounding box for the cylinder, then fully voxelize this bounding box in three directions, just like the cuboid blank. Finally, select the voxels belonging to the cylindrical part to obtain the voxel model of the cylindrical blank.

[0008] The voxels are categorized into three states, represented by three integers: 1, 0, and 2. 1 represents an effective voxel (containing material); 0 represents an empty voxel (formed after an effective voxel is cut); and 2 represents an invalid voxel (not recognized during simulation). For example, for a cylindrical blank, after the voxel model is built, voxels located within the rectangular bounding box of the workpiece but extending beyond the cylindrical blank's geometry are considered invalid voxels (the portion of the rectangular bounding box minus the cylindrical blank). Furthermore, for both rectangular and cylindrical blanks, voxels are sometimes converted to invalid voxels during simulation.

[0009] The states of all voxels in both voxel models are stored in a one-dimensional array, with each voxel having a unique index, as shown in Equation 1: (1) In the formula, , , The number of voxels in the X, Y, and Z directions of the rectangular or cylindrical bounding box; , , The numbers represent the voxels in the X, Y, and Z directions.

[0010] S2: Derive the maximum feed step size of the tool in three-axis machining mode to obtain the maximum feed step size of the two tools used at the interpolation tool position: A material removal method based on a three-dimensional distance field is used, employing discrete tool positions to control tool movement and perform cutting at each tool position. When the tool is performing five-axis simultaneous machining, the tool positions are dense and interpolation is unnecessary. However, when the tool is performing three-axis machining, such as when the tool tip moves linearly in the XOY plane, the input tool positions only contain the start and end coordinates of each line segment, requiring manual interpolation of intermediate tool positions. When the tool feed step size is set too large, the distance between two tool positions may become too great, potentially leading to incomplete voxel removal. Therefore, the maximum tool feed step size should be adjusted accordingly. The derivation process aims to remove as many voxels as possible from the toolpath while minimizing the number of tool positions after interpolation. This yields the minimum tool position coordinates required to meet simulation accuracy requirements, thus balancing accuracy and efficiency in the simulation process. Specifically: S2-1: Assume the tool moves along the X-axis in the XOY plane. , Let be the centers of the projection circles of the tool in the XOY plane at two adjacent tool positions. A voxel is a small cube, and its center is the geometric center of that cube. For any voxel center, the vector... A parallel straight line, when satisfying equation (2), allows voxels on the line to be completely and continuously removed.

[0011] (2) In the formula, This is the feed step of the tool, which is the distance between two tool positions; The distance vector between the centers of any voxels and the straight line. The distance; Where is the tool radius.

[0012] S2-2: To enable the tool to continuously remove all voxels along its trajectory, the following conditions must be met: (3) In the formula, Distance vector Farthest voxel center and vector The distance between them.

[0013] It is known that the tool cannot always move along the coordinate axes. Regardless of the direction the tool moves, All voxels removed can be found by finding the center and vector of all voxels removed at the first cutter point in the XOY projection plane. The distance between them is used to obtain the result. However, this would complicate the work and might encounter problems. The occurrence of this situation makes the calculated... If the value is extremely small, then equation (3) no longer applies. Therefore, to simplify the work, an empirical formula is needed to ensure that... It is readily available and versatile in most situations.

[0014] S2-3: When the tool moves along the X-axis in the XOY plane, we get: (4) in, Indicates the tool radius and The difference between them; Indicates voxel size; S2-4: For flat-end cylindrical cutters and ball-end cylindrical cutters, different methods should be selected based on their respective structural characteristics. Value. For flat-end cylindrical milling cutters, take... For ball-end mill cutters, since the diameter of the cutter varies in different sections of the ball end, a comprehensive consideration is required. Thus, by combining equations (3) and (4), the maximum feed step of the tool can be obtained. , The empirical formula for calculation is as follows: (5) In the formula, This is the maximum feed step of the flat-end cylindrical end mill; This represents the maximum feed step of the ball end mill.

[0015] S3: Establish a material removal method based on a three-dimensional distance field to achieve effective voxel removal during processing. Specifically: S3-1: Use the tool tip coordinates and tool axis rotation coordinates of the discrete tool position point to move the tool to the correct position.

[0016] S3-2: Transform the tool to the workpiece local coordinate system, construct the tool bounding box, and obtain the range of voxels indexed in the XYZ directions within the tool bounding box; S3-3: Traverse the voxels within the tool bounding box, calculate the 3D distance field for each valid voxel, and filter out voxels located within the tool geometry. Specifically: S3-3-1: For a flat-end mill on a five-axis machine tool, the following calculations are performed for each effective voxel within the tool enclosure. First, the coordinates of the voxel center are obtained: (6) in, , , Let the coordinates be the center coordinates of the voxels. Voxel size.

[0017] S3-3-2: Then, the vectorized projection parameters... Perform the calculation: (7) In the formula, The projection parameters are vectorized; the tool tip is used as the tool reference point. The vector formed by the center of the voxel in the workpiece coordinate system and the tool reference point; Let be the axial vector of the cutting edge portion of the tool in the workpiece coordinate system, and its length is the length of the cutting edge. .

[0018] S3-3-3: Then find the point on the tool axis that is closest to the voxel center. P p Coordinates: (8) In the formula, , , For point P p The coordinates; , , These are the coordinates of the tool reference point.

[0019] S3-3-4: The three-dimensional distance between each effective voxel within the tool bounding box and the tool. d The calculation is as follows: (9) In the formula, It is a three-dimensional distance; , , The coordinates of the voxel center; , , For point P p The coordinates.

[0020] S3-3-5: The condition that each effective voxel within the tool enclosure must satisfy for a flat-end mill to cut is: (10) In the formula, Where is the tool radius.

[0021] S3-3-6: For ball end mills on five-axis machine tools, the cutting edge consists of a cylindrical part and a ball end. The situation is the same when a voxel is cut by the cylindrical part of the cutting edge as with a flat end mill, but the cutting of the ball end needs to be considered separately. First, obtain the proportion of the ball end in the axial vector. ,in , Indicates the tool radius. This represents the length of the cutting edge of the tool. At this point, the projection parameters are satisfied. For voxels, the center of the sphere should be taken as the closest point of the tool axis for calculating the three-dimensional distance field; otherwise, it is impossible to effectively filter the voxels to be removed. Therefore, formula (11) is used to constrain the projection parameters of all voxels to [ Within the range of ,1].

[0022] (11) in, Represents the constrained vectorized projection parameters; S3-3-7: Then, equation (8) is transformed into equation (12): (12) S3-3-8: Then, use equation (9) to calculate the three-dimensional distance. Finally, the condition that a voxel must satisfy to be cut by a ball end mill on a five-axis machine tool is: (13) S3-4: Change the state of a voxel that meets the conditions of equation (10) or equation (13) from an effective voxel to an empty voxel.

[0023] S4: Establish a method for visualizing triangular meshes on the workpiece surface, thereby enabling the visualization and updating of the geometric surface of the workpiece during the machining process simulation: When visualizing the workpiece, all voxels are traversed to find their states. If the current voxel is a valid voxel, i.e., its state is "1", then the six orthogonal directions (up, down, left, right, front, back) of the voxel are further searched, and it is determined whether the states of the adjacent voxels in each direction are "1". If the state of the adjacent voxels in a certain direction is not "1", or there are no adjacent voxels, it means that the voxel is a surface voxel, and a surface triangular mesh of the voxel in the current direction is generated. The above operations are performed until the traversal is completed.

[0024] S5: Based on the four methods in S1 to S4, coupling and integration are carried out, and a static-dynamic voxel separation idea is proposed to form a material removal simulation method based on static-dynamic voxel separation, thereby realizing visualization and acceleration of the simulation process.

[0025] The Static-Dynamic Voxel Separation Method (SDVS) proposed in this invention accelerates visualization by dividing voxels in the machining simulation process into static and dynamic voxels. First, the material removal process is divided into several segments according to the actual machining steps, one segment per step. Then, a tool pre-movement process is added before each formal cutting simulation. During this pre-movement, the tool movement uses the same tool position control as in the formal cutting simulation. No visualization update is performed during the tool pre-movement, making this process much faster than the formal cutting simulation. Its purpose is to completely separate the voxels removed during the machining simulation (the remaining material in the actual machining) from the voxels remaining after the simulation (the part in the actual machining). The former is called the dynamic voxel, and the latter the static voxel. In this way, in the formal cutting simulation conducted after the tool pre-movement, only the dynamic voxels need to be updated, avoiding the repeated updates to the static voxels, thus reducing the visualization workload. The SDVS-based material removal simulation process can be divided into 7 steps: S5-1 initializes the voxel state using the workpiece modeling method based on uniform voxels proposed in S1. If necessary, the maximum tool feed step size in the three-axis machining mode is calculated in advance using the empirical formula for the maximum tool feed step size proposed in S2, and the tool position point is obtained by interpolation.

[0026] In S5-2, the tool pre-movement operation for the first machining step is performed, controlling the tool at the tool position point to quickly complete the material removal in the first stage of the machining process. Using the material removal method based on a three-dimensional distance field proposed in S3, the voxel state of the removed material changes from "1" to "0". No visualization updates are performed during this process.

[0027] S5-3, copy the voxel state array from S5-2 and change all "0"s to "2". This is the initialization for the next section of the cutting simulation.

[0028] In S5-4, for the voxels in S5-2, all voxels with a value of "1" are considered static voxels. The workpiece surface triangular mesh visualization method proposed in S4 is used to visualize the voxels, generating the surface triangular mesh of the first simulated static voxel.

[0029] In S5-5, the states of all voxels with a value of "1" and those with a value of "0" are swapped, while voxels with a value of "2" are ignored. At this point, all voxels with a value of "1" are dynamic voxels. The workpiece surface triangular mesh visualization method proposed in S4 is used again to visualize the dynamic voxels of the first simulation, generating the surface triangular mesh of the dynamic voxels of the first simulation.

[0030] In S5-6, the formal cutting simulation operation of the first process is executed. The tool moves again according to the tool position point, and the material removal method based on the three-dimensional distance field proposed in S3 is used to complete the first cutting process. After each material removal, the surface triangular mesh of the dynamic voxels is updated again using the workpiece surface triangular mesh visualization method proposed in S4, until all dynamic voxels are cut, that is, all "1"s become "0".

[0031] S5-7: After completing the initialization of the voxel state in the second stage of simulation in S5-3, the second stage of tool pre-movement operation and voxel state transformation are performed on this basis.

[0032] S5-8, repeat steps S5-2 to S5-7 until the entire cutting simulation process is completed, finally obtaining the geometric model of the workpiece after machining, and verifying the correctness of the toolpath.

[0033] The beneficial effects of this invention are: (1) This invention addresses the problem of low visualization efficiency in traditional voxel-based material removal simulation of machining processes. In the field of rapid simulation of machining processes, it proposes a visualization acceleration method for simulating the machining of complex workpieces with large material removal volumes. This method is based on static-dynamic voxel separation, which can divide the simulation process into several segments and perform tool pre-movement and formal cutting simulation for each segment in sequence. This sequentially separates the dynamic voxels of each segment from all voxels, achieving dual decoupling of static and dynamic voxels in time and space. This effectively controls the number of dynamic voxels in each formal cutting simulation segment, thereby significantly reducing the visualization workload of the simulation process.

[0034] (2) The present invention proposes an empirical formula for the maximum feed step of the tool in the three-axis machining mode, which can minimize the number of tool positions when interpolating to obtain the tool position.

[0035] In summary, the method of the present invention can solve the problem of low visualization efficiency in traditional voxel-based material removal simulation of machining processes, and realize rapid simulation of machining processes. Attached Figure Description

[0036] Figure 1 Auxiliary diagram for deriving the maximum feed step of the tool in three-axis machining mode; Figure 2 Flowchart of a method for visualizing triangular meshes on workpiece surfaces; Figure 3 This is a schematic diagram of the simulation process for material removal based on static-dynamic voxel separation; Figure 4 This is a simulation diagram of the processing in Example 1; Figure 5 Comparison of computational efficiency between SDVS and traditional methods; Figure 5 (a) in the figure represents a comparison of calculation times; Figure 5 (b) represents the time reduction rate of SDVS compared to the traditional method; Figure 6 These are simulation diagrams of the manufacturing process for three complex embodiments; Figure 6 (a) in the figure is a simulation diagram of the blade manufacturing process; Figure 6 (b) in the diagram is a simulation of the impeller machining process; Figure 6 (c) in the figure is a simulation diagram of the bladed disk manufacturing process; Figure 7 This is a flowchart of the method of the present invention. Detailed Implementation

[0037] To make the technical solution and advantages of the present invention clearer, a material removal simulation method based on static-dynamic voxel separation is introduced below to test the simulation speed-up effect of Example 1. Figure 4 As shown, in Example 1, 24 identical grooves were arranged on a 60×60×10mm rectangular blank, and machined using a 4mm diameter flat-end mill D4. (Voxenon dimensions) s The total number of elements is 0.15 mm. N v The value is 10,720,000. The amount of material removed is controlled by the number of grooves (1-24) cut by the tool in a single operation, with each groove being a separate cutting process. The technical solution of the present invention will be clearly and completely described in conjunction with specific embodiments and accompanying drawings.

[0038] This embodiment addresses the low visualization efficiency of traditional voxel-based simulations of material removal in machining processes. In the field of rapid simulation of machining processes, it proposes a visualization acceleration method for simulating complex workpieces with large material removal volumes. The method flow is as follows: Figure 7 As shown. This embodiment implements a material removal simulation method based on static-dynamic voxel separation. First, a workpiece modeling method based on uniform voxels is established; second, the maximum tool feed step size in three-axis machining mode is derived, and the tool position point is obtained through interpolation; third, a material removal method based on a three-dimensional distance field is established; fourth, a workpiece surface triangular mesh visualization method is established; fifth, a static-dynamic voxel separation method is established, and the previous methods are integrated to form the final proposed material removal process simulation acceleration method. This method is used to perform rapid simulation of the embodiment. The steps include: S1: Establish a workpiece modeling method based on uniform voxels to obtain discretized voxel models of classic cuboid and cylindrical blanks. Specifically: To model the cuboid blank geometry used in this embodiment using voxels of uniform size, it is necessary to input its length and voxel dimensions in the X, Y, and Z directions. The model is discretized and voxelized in three directions to obtain a voxel model of a cuboid blank. Each voxel has three states, represented by three integers: 1, 0, and 2. 1 represents an effective voxel (containing material); 0 represents an empty voxel (the effective voxel after being cut); and 2 represents an ineffective voxel (not recognized during simulation).

[0039] The states of all voxels in the voxel model are stored in a one-dimensional array, with each voxel having a unique index, as shown in Equation 1: (1) In the formula, , , The number of voxels in the X, Y, and Z directions of the rectangular or cylindrical bounding box; , , The numbers represent the voxels in the X, Y, and Z directions.

[0040] S2: The maximum feed step size of the tool in the three-axis machining mode is derived to obtain the maximum feed step size of the two tools used at the interpolation tool position. Specifically: S2-1: Assume the tool moves along the X-axis in the XOY plane, such as Figure 1 As shown, , Let be the centers of the projection circles of the tool in the XOY plane at two adjacent tool positions. A voxel is a small cube, and its center is the geometric center of that cube. For any voxel center, the vector... A parallel straight line, when satisfying equation (2), allows voxels on the line to be completely and continuously removed.

[0041] (2) In the formula, This is the feed step of the tool, which is the distance between two tool positions; The distance vector between the centers of any voxels and the straight line. The distance; Where is the tool radius.

[0042] S2-2: To enable the tool to continuously remove all voxels along its trajectory, the following conditions must be met: (3) In the formula, Distance vector Farthest voxel center and vector The distance between them.

[0043] It is known that the tool cannot always move along the coordinate axes. Regardless of the direction the tool moves, All voxels removed can be found by finding the center and vector of all voxels removed at the first cutter point in the XOY projection plane. The distance between them is used to obtain the result. However, this would complicate the work and might encounter problems. The occurrence of this situation makes the calculated... If the value is extremely small, then equation (3) no longer applies. Therefore, to simplify the work, an empirical formula is needed to ensure that... It is readily available and versatile in most situations.

[0044] S2-3: When the tool moves along the X-axis in the XOY plane, we get: (4) in, Indicates the tool radius and The difference between them; Indicates voxel size; S2-4: For flat-end cylindrical cutters and ball-end cylindrical cutters, different methods should be selected based on their respective structural characteristics. Value. For flat-end cylindrical milling cutters, take... For ball-end mill cutters, since the diameter of the cutter varies in different sections of the ball end, a comprehensive consideration is required. Thus, by combining equations (3) and (4), the maximum feed step of the tool can be obtained. , The empirical formula for calculation is as follows: (5) In the formula, This is the maximum feed step of the flat-end cylindrical end mill; This represents the maximum feed step of the ball end mill.

[0045] The cutting tool used in this embodiment is a flat-end cylindrical end mill D4. =2mm, =0.15mm, therefore we can obtain =0.69mm.

[0046] S3: Establish a material removal method based on a three-dimensional distance field to achieve effective voxel removal during processing. Specifically: S3-1: Use the tool tip coordinates and tool axis rotation coordinates of the discrete tool position point to move the tool to the correct position.

[0047] S3-2: Transform the tool to the workpiece local coordinate system, construct the tool bounding box, and obtain the range of voxels indexed in the XYZ directions within the tool bounding box; S3-3: Traverse the voxels within the tool bounding box, calculate the 3D distance field for each valid voxel, and filter out voxels located within the tool geometry. Specifically: S3-3-1: For a flat-end mill on a five-axis machine tool, the following calculations are performed for each effective voxel within the tool enclosure. First, the coordinates of the voxel center are obtained: (6) in, , , Let the coordinates be the center coordinates of the voxels. Voxel size.

[0048] S3-3-2: Then, the vectorized projection parameters... Perform the calculation: (7) In the formula, The projection parameters are vectorized; the tool tip is used as the tool reference point. The vector formed by the center of the voxel in the workpiece coordinate system and the tool reference point; Let be the axial vector of the cutting edge portion of the tool in the workpiece coordinate system, and its length is the length of the cutting edge. .

[0049] S3-3-3: Then find the point on the tool axis that is closest to the voxel center. P p Coordinates: (8) In the formula, , , For point P p The coordinates; , , These are the coordinates of the tool reference point.

[0050] S3-3-4: The three-dimensional distance between each effective voxel within the tool bounding box and the tool. d The calculation is as follows: (9) In the formula, It is a three-dimensional distance; , , The coordinates of the voxel center; , , For point P p The coordinates.

[0051] S3-3-5: The condition that each effective voxel within the tool enclosure must satisfy for a flat-end mill to cut is: (10) In the formula, Where is the tool radius.

[0052] S3-4: Change the state of a voxel that meets the conditions of equation (10) or equation (13) from an effective voxel to an empty voxel.

[0053] S4: Establish a method for visualizing triangular meshes on the workpiece surface, thereby enabling the visualization and updating of the geometric surface of the workpiece during the machining process simulation: The flowchart of the workpiece surface triangular mesh visualization method is as follows: Figure 2 As shown. When visualizing the workpiece, all voxels are traversed to find their states. If the current voxel is a valid voxel, i.e., its state is "1", then the six orthogonal directions (up, down, left, right, front, back) of the voxel are further searched, and it is determined whether the states of the adjacent voxels in each direction are "1". If the state of the adjacent voxels in a certain direction is not "1", or there are no adjacent voxels, it means that the voxel is a surface voxel, and a surface triangular mesh of the voxel in the current direction is generated. The above operations are performed until the traversal is completed.

[0054] S5: Based on the four methods in S1 to S4, a coupling and integration method is proposed, and a static-dynamic voxel separation approach is developed to form a material removal simulation method based on static-dynamic voxel separation, thereby achieving visualization and acceleration of the simulation process. Visualization acceleration is achieved by dividing the voxels in the machining simulation process into static and dynamic voxels. First, the material removal process is divided into several segments according to the actual machining steps, one segment per step. Then, a tool pre-movement process is added before each formal cutting simulation. During this pre-movement, the tool movement uses the same tool position control as in the formal cutting simulation. No visualization updates are performed during the tool pre-movement, making this process much faster than the formal cutting simulation. Its purpose is to completely separate the voxels removed during the machining simulation (the remaining material in the actual machining) from the voxels remaining after the simulation (the actual machined part). The former are called dynamic voxels, and the latter are called static voxels. In this way, in the formal cutting simulation following the tool pre-movement, only the dynamic voxels need to be updated, avoiding the repetitive updates to the static voxels, thus reducing the visualization workload. The SDVS material removal simulation process can be divided into 7 steps, as illustrated in the flowchart below. Figure 3 As shown: S5-1 initializes the voxel state using the workpiece modeling method based on uniform voxels proposed in S1. If necessary, the maximum tool feed step size in the three-axis machining mode is calculated in advance using the empirical formula for the maximum tool feed step size proposed in S2, and the tool position point is obtained by interpolation.

[0055] In S5-2, the tool pre-movement operation for the first machining step is performed, controlling the tool at the tool position point to quickly complete the material removal in the first stage of the machining process. Using the material removal method based on a three-dimensional distance field proposed in S3, the voxel state of the removed material changes from "1" to "0". No visualization updates are performed during this process.

[0056] S5-3, copy the voxel state array from S5-2 and change all "0"s to "2". This is the initialization for the next section of the cutting simulation.

[0057] In S5-4, for the voxels in S5-2, all voxels with a value of "1" are considered static voxels. The workpiece surface triangular mesh visualization method proposed in S4 is used to visualize the voxels, generating the surface triangular mesh of the first simulated static voxel.

[0058] In S5-5, the states of all voxels with a value of "1" and those with a value of "0" are swapped, while voxels with a value of "2" are ignored. At this point, all voxels with a value of "1" are dynamic voxels. The workpiece surface triangular mesh visualization method proposed in S4 is used again to visualize the dynamic voxels of the first simulation, generating the surface triangular mesh of the dynamic voxels of the first simulation.

[0059] In S5-6, the formal cutting simulation operation of the first process is executed. The tool moves again according to the tool position point, and the material removal method based on the three-dimensional distance field proposed in S3 is used to complete the first cutting process. After each material removal, the surface triangular mesh of the dynamic voxels is updated again using the workpiece surface triangular mesh visualization method proposed in S4, until all dynamic voxels are cut, that is, all "1"s become "0".

[0060] S5-7: After completing the initialization of the voxel state in the second stage of simulation in S5-3, the second stage of tool pre-movement operation and voxel state transformation are performed on this basis.

[0061] S5-8, repeat steps S5-2 to S5-7 until the entire cutting simulation process is completed, finally obtaining the geometric model of the workpiece after machining, and verifying the correctness of the toolpath.

[0062] After the simulation is complete, you can... Figure 5 The acceleration effect of the proposed method under different material removal rates was obtained. Figure 5 As can be seen in (a) of the diagram, under the same material removal rate, the computation time of SDVS is significantly shorter than that of traditional methods. Figure 5 As can be seen in (b), as the amount of material removed increases, the time reduction rate of SDVS compared to the traditional method gradually increases, with an overall time reduction rate of 59%~72%, proving that it is suitable for processing simulation with a large amount of material removal.

[0063] To test the acceleration effect of the proposed method in the simulation of machining complex workpieces with large material removal volumes, this invention provides three additional embodiments, which simulate the machining of complex thin-walled workpieces such as blades, impellers, and impeller disks. Information for these three embodiments is shown in Table 1. The total number of voxels reaches tens of millions, and the number of discrete tool points reaches hundreds of thousands; such a scale of computation cannot be accomplished by traditional methods. Figure 6 The document demonstrates part of the simulation process and the final result of applying SDVS in three embodiments. Figure 6 (a) in the figure is a simulation diagram of the blade manufacturing process; Figure 6 (b) in the diagram is a simulation of the impeller machining process; Figure 6 (c) in the diagram shows the simulation process of the bladed disk machining. It can be seen that the static and dynamic voxels in each process are completely separated. This not only effectively reduces the amount of visualization work but also clearly shows the location of the material removed in each process. If overcutting or undercutting occurs, the user can more quickly locate the problematic process.

[0064] Table 1: Information on Complex Implementation Examples

[0065] The simulation results of the machining process of a specific workpiece show that, compared with traditional methods, the static-dynamic voxel separation method proposed in this application can significantly reduce the simulation calculation time and can complete simulation tasks with tens of millions of voxels, which are difficult to achieve with traditional methods. Therefore, the method proposed in this application is applicable to the machining simulation tasks of complex workpieces with large amounts of material removal.

[0066] The above-described embodiments are merely illustrative of the implementation methods of the present invention, but should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the protection scope of the present invention.

Claims

1. A visualization-based accelerated simulation method for machining complex workpieces with large material removal volumes, characterized in that, The visualization acceleration method includes the following steps: S1: Establish a workpiece modeling method based on uniform voxels to obtain discretized voxel models of classic cuboid and cylindrical blanks; specifically: Using voxels of the same size, the cuboid blank and the cylindrical blank in the five-axis machining are modeled separately to obtain the voxel model of the cuboid blank and the voxel model of the cylindrical blank. The state of all voxels in the two voxel models is stored in a one-dimensional array, and each voxel has a unique index. S2: The maximum feed step of the tool in the three-axis machining mode is derived. Discrete tool positions are obtained through interpolation. The maximum feed step of the two tools used when interpolating the tool positions is obtained. A material removal method based on a three-dimensional distance field is used, which controls tool movement using discrete tool positions and performs cutting at each tool position. When the tool is performing five-axis simultaneous machining, tool position interpolation is not required; however, when the tool is performing three-axis machining, interpolation is required at intermediate tool positions. The maximum feed step size of the tool is also considered. The derivation process removes all voxels from the toolpath and reduces the number of tool points after interpolation, resulting in the minimum tool point coordinates that meet the simulation accuracy requirements. S3: Establish a material removal method based on a three-dimensional distance field to achieve effective voxel removal during processing; specifically: S3-1: Use the tool tip coordinates and tool axis rotation coordinates of the discrete tool position point to move the tool to the correct position; S3-2: Transform the tool to the workpiece local coordinate system, construct the tool bounding box, and obtain the range of voxels indexed in the XYZ directions within the tool bounding box; S3-3: Traverse the voxels within the tool bounding box, calculate the three-dimensional distance field of each valid voxel, and filter out the voxels located inside the tool geometry. S3-4: Change the status of eligible voxels from valid voxels to empty voxels; S4: Establish a method for visualizing triangular meshes on the workpiece surface to achieve visual updates of the geometric surface of the workpiece during machining simulation. When visualizing the workpiece, all voxels are traversed to find the voxel state. If the current voxel is a valid voxel, i.e., the voxel state is 1, then the six orthogonal directions of the current voxel are further searched, and it is determined whether the state of the adjacent voxels in each direction is 1. If the state of the adjacent voxels of a voxel in a certain direction is not 1, or there are no adjacent voxels, it means that the current voxel is a surface voxel, and then the surface triangular mesh of the current voxel in the current direction is generated. This process continues until the traversal is completed. S5: Based on the coupling integration of S1~S4, a static-dynamic voxel separation method is established, forming a material removal simulation method based on static-dynamic voxel separation, thereby realizing visualization and acceleration of the simulation process; Visualization acceleration is achieved by dividing the voxels in the machining simulation process into static and dynamic voxels. First, the material removal process is divided into several segments according to the actual machining steps, with each step consisting of one segment. Then, a tool pre-movement process is added before each formal cutting simulation. During the tool pre-movement, the tool movement is controlled by the same tool position point as in the formal cutting simulation. No visualization update is performed during the tool pre-movement, and the voxels removed during the machining simulation are completely separated from the voxels remaining after the simulation. The former is called dynamic voxels, and the latter is called static voxels. In the formal cutting simulation that follows the tool pre-movement, only the dynamic voxels are visualized and updated.

2. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 1, characterized in that, In S1: For a cuboid blank, input its length and voxel dimensions in the X, Y, and Z directions. Discretize and voxelize in three directions to obtain the voxel model of the cuboid blank; for the cylindrical blank, after inputting the diameter, axial length, central axis direction and voxel size, first establish the cuboid bounding box of the cylinder, fully voxelize the bounding box in three directions, and finally filter out the voxels belonging to the cylindrical part to obtain the voxel model of the cylindrical blank. The voxels are classified into three states, represented by three integers: 1, 0, and 2. 1 represents an effective voxel, which is a voxel containing material; 0 represents an empty voxel, which is a voxel obtained after the effective voxel is cut; and 2 represents an invalid voxel, which is a voxel not recognized during the simulation. For both cuboid and cylindrical blanks, voxels are transformed into invalid voxels during the simulation. Each voxel has a unique index as shown in Equation 1: (1) In the formula, , , The number of voxels in the X, Y, and Z directions of the rectangular or cylindrical bounding box; , , The numbers represent the voxels in the X, Y, and Z directions.

3. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 2, characterized in that, Specifically, S2 is: S2-1: Assume the tool moves along the X-axis in the XOY plane. , Let be the centers of the projection circles of the tool in the XOY plane at two adjacent tool positions; let voxels be small cubes, and let the center of each voxel be the geometric center of the small cube; for any voxel center, let the vector be... Parallel straight lines, when satisfying equation (2), have voxels completely and continuously removed; (2) In the formula, This is the feed step of the tool, which is the distance between two tool positions; The distance vector between the centers of any voxels and the straight line. The distance; The radius of the cutting tool; S2-2: To enable the tool to continuously remove all voxels along its trajectory, the following conditions must be met: (3) In the formula, Distance vector Farthest voxel center and vector The distance between; S2-3: When the tool moves along the X-axis in the XOY plane, we get: (4) in, Indicates tool radius and The difference between them; Indicates voxel size; S2-4: For flat-end cylindrical cutters and ball-end cylindrical cutters, different methods should be selected based on their respective structural characteristics. Value; for flat-end cylindrical milling cutters For ball-end mill cutters, take Combining equations (3) and (4), the maximum feed step of the tool is obtained. , The empirical formula for calculation is as follows: (5) In the formula, This is the maximum feed step of the flat-end cylindrical end mill; This represents the maximum feed step of the ball end mill.

4. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 3, characterized in that, Specifically, S3-3 is: S3-3-1: For a flat-end mill cutter on a five-axis machine tool, for each effective voxel within the tool enclosure, first obtain the coordinates of the voxel center: (6) in, , , Let the coordinates be the center coordinates of the voxel. Voxel size; S3-3-2: Vectorized projection parameters Perform the calculation: (7) In the formula, The projection parameters are vectorized; the tool tip is used as the tool reference point. The vector formed by the center of the voxel in the workpiece coordinate system and the tool reference point; Let be the axial vector of the cutting edge portion of the tool in the workpiece coordinate system, and its length is the length of the cutting edge. ; S3-3-3: Find the point on the tool axis closest to the voxel center. P p Coordinates: (8) In the formula, , , For point P p The coordinates; , , The coordinates of the tool reference point; S3-3-4: The three-dimensional distance between each effective voxel within the tool enclosure and the tool. d The calculation is as follows: (9) In the formula, It is a three-dimensional distance; , , The coordinates of the voxel center; , , For point P p The coordinates; S3-3-5: The condition that each effective voxel within the tool enclosure must satisfy for a flat-end mill to cut is: (10) In the formula, The radius of the cutting tool; S3-3-6: For ball end mills on five-axis machine tools, the cutting edge is divided into a cylindrical part and a ball end part; the situation when a voxel is cut by the cylindrical part of the cutting edge is the same as that of a flat end mill, but the cutting of the ball end part needs to be considered separately; first, obtain the proportion of the ball end part in the axial vector. ,in , Indicates the tool radius. This indicates the length of the cutting edge of the tool; at this point, the projection parameters are satisfied. For voxels, the center of the sphere should be taken as the nearest point of the tool axis for calculating the three-dimensional distance field; otherwise, the voxels to be removed cannot be effectively screened. Therefore, formula (11) is used to constrain the projection parameters of all voxels within [ Within the range of ,1]; (11) in, Represents the constrained vectorized projection parameters; S3-3-7: Then, equation (8) is transformed into equation (12): (12) S3-3-8: Then, use equation (9) to calculate the three-dimensional distance. Finally, the condition that a voxel must satisfy to be cut by a ball end mill on a five-axis machine tool is: (13)。 5. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 4, characterized in that, In S3-4, the state of a voxel that meets the conditions of formula (10) or formula (13) is changed from an effective voxel to an empty voxel.

6. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 5, characterized in that, In S4, the six orthogonal directions refer to the current voxel's up, down, left, right, front, and back directions.

7. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 6, characterized in that, Specifically, S5 is: S5-1, the workpiece modeling method based on uniform voxels proposed in S1 is used to initialize the voxel state; S5-2, the tool pre-movement operation for the first operation is performed, and the tool is controlled by the tool position point to quickly complete the material removal of the first stage of the machining process; using the material removal method based on the three-dimensional distance field proposed in S3, the voxel state of the removed material changes from 1 to 0; no visualization update is performed during this process; S5-3, copy the voxel state array of S5-2 and change all 0s to 2s; this is the initialization for the next section of the cutting simulation. S5-4, for the voxels in S5-2, all voxels with a value of 1 are static voxels; The workpiece surface triangular mesh visualization method proposed by S4 is used to visualize the voxels and generate the surface triangular mesh of the first simulated static voxel. S5-5, swap the states of all voxels with a state of 1 and voxels with a state of 0, and ignore voxels with a state of 2; at this time, all voxels with a state of 1 are dynamic voxels. Visualize the dynamic voxels of the first simulation again and generate the surface triangular mesh of the dynamic voxels of the first simulation. S5-6, execute the formal cutting simulation operation for the first process. The tool moves again according to the tool position point to complete the first cutting process. After each material removal, the surface triangular mesh of the dynamic voxels is updated again until all dynamic voxels are cut off, i.e. all 1s become 0s. S5-7: The voxel state initialization of the second stage of simulation has been completed in S5-3. Based on this, the second stage of tool pre-movement operation and voxel state transformation are performed. S5-8, repeat steps S5-2 to S5-7 until the entire cutting simulation process is completed and the geometric model of the workpiece after machining is obtained.

8. The visualization-based acceleration method for simulating the machining of complex workpieces with large material removal volumes according to claim 7, characterized in that, In S5-1, the maximum feed step length of the tool in the three-axis machining mode is calculated in advance using the empirical formula for the maximum feed step length of the tool proposed in S2, and the tool position point is obtained by interpolation.

Images