Method for generating a machining path of a free-form surface based on reparameterization

Abstract

The application discloses a kind of based on reparameterization's cutting free surface equal parameter processing path generation method.The application is first to the entity model of import pre-processing, mainly to the B-rep entity model of complex cutting surface composition is decut, so that it is in the new B-rep entity model of multiple complete tensor product surface composition by surface reparameterization conversion within the given accuracy after reconstruction face and face adjacent relationship, the whole model is segmented into multiple surface zone;Finally, each surface on the surface zone is generated equal parameter line processing track one by one.The application can be converted into the model of watertight complete surface composition within the accuracy requirement set by user by cutting surface composition model, and generates the numerical control processing track of model, avoids the gap between cutting surface model to make processing track cannot be continuously generated and the difficult problem of low processing efficiency, realizes efficient, robust processing path generation.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of Computer-Aided Design and Manufacturing (CAD / CAM) system, and particularly relates to a machining path generation and product manufacturing method based on surface reparameterization. BACKGROUND

[0002] In the process of numerical control machining, the tool machining path determines the shape accuracy and surface roughness of the machined part, and the planning result of the tool machining trajectory is largely determined by the model shape. Therefore, the quality of the model determines the quality of the final product, and the quality of the model cannot be separated from the solid modeling method in the CAD / CAM system.

[0003] Boolean operation is one of the foundations of solid modeling methods, and early research on Boolean operation focuses on the robust Boolean method between geometric bodies. This establishes the paradigm of using Boolean operation as a basic tool to create complex geometric bodies from simple geometric bodies in a hierarchical manner. Because of the strong expression ability of NURBS system, most CAD / CAM systems use NURBS surfaces when using boundary representation. Therefore, the classical solid model Boolean operation problem becomes the problem of calculating the intersection of two parameter surfaces. This face-face intersection problem has been proven to be difficult to obtain a closed-form solution in mathematics, which leads to a large number of research and use of numerical approximation methods.

[0004] The numerical approximation system method used for cutting surfaces brings many problems to the usability of the model, among which the most difficult one is the watertightness of the model. Because the approximate representation of the same intersection line on the parameter domain of the two intersecting parameter surfaces cannot exactly coincide without error, it can describe the same curve in the model space, so there will always be a small "crack" or "bump" at the intersection line in the design intent. This is the "non-watertightness" of the model. The "crack" or "bump" is too small to be visible to the human eye, but it cannot be ignored for CAE analysis software and numerical control machines. Its existence seriously affects the robustness of various geometric methods. Especially for cutting surfaces with holes, numerical non-robustness may cause the machining trajectory to enter the forbidden zone, resulting in defects or even damage to the finished machining model.

[0005] In addition, from the perspective of processing efficiency, for the curved surface with holes, in order to make the added trajectory not enter the forbidden zone, the machine tool usually needs to use G00 mode to make the tool pass through the forbidden zone from a far enough position above it, which drags the processing efficiency. If the number of holes is large and the shape is complex, it is a great burden to the processing efficiency. The trend of modern machining is to shorten the processing time of the model through reasonable planning of the processing trajectory, and to calculate the efficiency of the machining. Obviously, the processing method of processing the trimmed surfaces in the model one by one cannot meet the requirement of high efficiency, and a new idea of processing is needed to plan the processing trajectory from the whole object level, and to improve the efficiency of numerical control machining as much as possible.

[0006] In summary, the model composed of trimmed surfaces is not friendly to the numerical control machining equipment, on the one hand, there is the problem of model quality caused by numerical robustness, and on the other hand, there is the problem of low processing efficiency caused by the processing method. SUMMARY

[0007] In view of the deficiencies in the background art, the purpose of the present application is to provide a complex trimmed free-form surface parameter machining path design and manufacturing method based on surface reparameterization.

[0008] Under the condition that the user imports a solid model through a CAD / CAM system and sets given machining parameters (mainly including numerical control machine tool type, tool size, machining accuracy, etc.), the machining path of the solid model composed of complex trimmed surfaces can be generated, and effective machining instructions can be generated for manufacturing.

[0009] The products described in the present application are high-end manufacturing products.

[0010] To achieve the above purpose, the technical scheme adopted by the present application is as follows:

[0011] 1) After reconstructing the B-rep solid model composed of trimmed surfaces, a B-rep solid model composed of multiple complete tensor product surfaces is obtained and is recorded as a target B-rep solid model, in the target B-rep solid model, adjacent tensor product surfaces are water-tightly connected, and two adjacent tensor product surfaces have the same node vector along the intersection line direction;

[0012] 2) According to the topological connection relationship between the tensor product surfaces in the target B-rep solid model, the target B-rep solid model is divided into multiple tensor product surface strips;

[0013] 3) The equal parameter line machining trajectories of all the tensor product surfaces on each tensor product surface strip are generated respectively, so as to obtain the machining path of the whole model.

[0014] The 1) is specifically:

[0015] 1.1) Divide the clipping domain of each clipping surface into multiple curved quadrilateral regions to obtain the initial surface corresponding to the current clipping domain. Traverse all clipping surfaces in the B-rep solid model composed of clipping surfaces to obtain the initial surface corresponding to all clipping surfaces, thereby obtaining the updated B-rep solid model.

[0016] 1.2) Fit the new intersection lines between adjacent initial surfaces in the updated B-rep solid model, and then extract the isoparametric lines of each initial surface based on the new intersection lines;

[0017] 1.3) By reparameterizing the isoparameter lines of each initial surface, the control points of each initial surface are obtained, and then each tensor product surface is obtained. Thus, a B-rep solid model composed of multiple complete tensor product surfaces is formed and denoted as the target B-rep solid model.

[0018] In section 1.1), the specific steps for processing the initial surface of each trimmed surface are as follows:

[0019] 1.1.1) Break all clipping curves within the current clipping domain at all non-differentiable points to obtain multiple smooth curves;

[0020] 1.1.2) Along the feature points in the smooth curve, divide all the smooth curves into corresponding sub-segments, and then divide each sub-segment into the main u feature segment and the main v feature segment to obtain the feature sub-segment set;

[0021] 1.1.3) Extract a sub-region in the current clipping domain based on the main extension direction of each feature segment in the feature segment set, so that each feature segment is a partial boundary of the sub-region. Traverse the feature segment set and use all feature segments as partial boundaries of the corresponding sub-regions to obtain the preliminary segmentation result of the current clipping domain.

[0022] 1.1.4) Based on the preliminary segmentation results of the current clipping domain, the current clipping domain is further divided into several quadrilateral regions to obtain the initial surface corresponding to the current clipping domain.

[0023] As stated in 1.2), the generation of isoparametric lines for each initial surface specifically includes the following steps:

[0024] 1.2.1) After fitting scattered points at the non-watertight intersection of two adjacent initial surfaces, a definite curve is generated and used as the new intersection line;

[0025] 1.2.2) Extract multiple isoparametric lines from each trimmed surface and trim or extend multiple isoparametric lines so that the endpoints of multiple isoparametric lines fall on the current new intersection line, thereby obtaining multiple processed isoparametric lines and using them as the isoparametric lines of the corresponding initial surface in the parameter domain;

[0026] 1.2.3) After transforming the isoparametric lines of the initial surface in the parameter domain to the model space, the isoparametric lines of the initial surface in the model space are obtained.

[0027] In section 1.3), the generation of the tensor product surface for each initial surface specifically includes the following steps:

[0028] 1.3.1) Perform a reparameterization transformation on the isoparameter lines corresponding to each initial surface in the model space so that the starting parameter value and the ending parameter value of each isoparameter line are the same, and obtain the transformed isoparameter lines;

[0029] 1.3.2) By utilizing the linear relationship between the control points of the B-spline surface and the control points of the transformed isoparametric lines, the control point mesh of the initial surface in the model space is obtained;

[0030] 1.3.3) Assign the control points of the new intersection line of the current initial surface to the control points on the edge of the control point grid of the initial surface in the model space, so that the B-spline surface in the control point grid is interpolated at the new intersection line.

[0031] 1.3.4) Use the node vectors and orders of the new intersection line as the node vectors and orders of the initial surface in the direction of the new intersection line, and use the node vectors and orders of the transformed isoparametric lines as the node vectors and orders of the initial surface in the direction of the isoparametric lines, thereby obtaining the tensor product surface corresponding to the current initial surface.

[0032] Specifically, 2) refers to:

[0033] 2.1) Construct two topological relationship tables based on the topological relationships of the target B-rep entity model. The two topological relationship tables are the adjacent surface table and the opposite edge relationship table, respectively.

[0034] 2.2) Based on the adjacent surface table and the opposite edge relation table, use the traversal method to generate multiple tensor product surface bands of the target B-rep solid model.

[0035] Specifically, 2.2) refers to:

[0036] 2.2.1) The first set S is composed of all tensor product surfaces in the target B-rep entity model, and then two surface lists L1 and L2 are initialized with two empty lists;

[0037] 2.2.2) Select any tensor product surface A and a pair of its opposite edges e1 and e2 from the current first set S, and add the tensor product surface A to the first surface strip list L1;

[0038] 2.2.3) Obtain the tensor product surface B adjacent to edge e1 of tensor product surface A from the adjacent surface table, add tensor product surface B to the first surface patch list L1, and obtain the opposite edge e3 of edge e1 on tensor product surface B from the opposite edge relationship table. Then, find the adjacent surface of tensor product surface B with respect to edge e3 and add it to the first surface patch list L1. Continue until the edge of the target B-rep solid model is reached to obtain the final first surface patch list L1.

[0039] 2.2.4) If edge e2 is not at the boundary of the target B-rep solid model, find the tensor product surface C of the tensor product surface A with respect to edge e2 from the adjacent surface table, add the tensor product surface C to the second surface strip list L2, find the opposite edge e4 of edge e2 on the tensor product surface C from the opposite edge relationship table, find the tensor product surface C with respect to edge e4 and add it to the second surface strip list L2, until the edge of the target B-rep solid model is reached, and obtain the final second surface strip list L2;

[0040] 2.2.5) Reverse the final second surface strip list L2 and then concatenate it with the final first surface strip list L1 to generate a tensor product surface strip and save it. Then, delete all tensor product surfaces of the tensor product surface strip from the first set S and update the first set S.

[0041] 2.2.6) Repeat steps 2.2.1)-2.2.5) until the first set S becomes an empty set, and obtain multiple tensor product surface bands.

[0042] Specifically, 3) refers to:

[0043] 3.1) Use path planning methods to generate tool points on all tensor product surface bands;

[0044] 3.2) Convert the tool position information into G-code instructions and use them for CNC machining.

[0045] Specifically, 3.1.1) refers to:

[0046] 3.1.1) Initialize the parameter values ​​in the direction of the new intersection line;

[0047] 3.1.2) Generate tool positions on the isoparameter lines corresponding to the parameter values ​​in the direction of the new intersection line on the entire tensor product surface band according to the machining accuracy in the parameter domain;

[0048] 3.1.3) Move the tool position point in the direction of increasing parameter value in the new intersection line direction according to the set residual height allowable value. Then take the tool position point with the smallest increment in the new intersection line direction among the tool positions on the entire curved surface as the auto-increment value of parameter value in the new intersection line direction, and update the parameter value in the new intersection line direction.

[0049] 3.1.4) Based on the parameter values ​​in the direction of the new intersection line, repeat steps 3.1.2)-3.1.3) until all tool points of the entire tensor product surface band are generated, thus obtaining all tool points of the current tensor product surface band.

[0050] After the user imports the model through the CAD / CAM system interface, based on the set machining accuracy and tool radius, this invention can generate isoparametric machining paths for solid models composed of trimmed surfaces, avoiding a series of problems such as robustness and inefficiency caused by the traditional isoparametric machining paths needing to be truncated and reconstructed by trimmed curves.

[0051] The beneficial effects of this invention are:

[0052] This invention first performs de-clipping and reconstruction on the model composed of clipped surfaces, transforming all surfaces into watertight, unclipping complete tensor product surfaces, and making the node vectors of adjacent surfaces homogeneous along the intersection direction, naturally generating isoparametric processing trajectories. The natural transition of isoparametric trajectories between adjacent surfaces reduces the use of G00 mode, which can effectively improve processing efficiency.

[0053] This invention can completely eliminate many difficulties caused by cutting curves before generating the trajectory, making the generation of the processing trajectory more robust and the processing efficiency higher. Attached Figure Description

[0054] Figure 1 This is the overall flowchart of the present invention.

[0055] Figure 2 This is a diagram illustrating the method for generating surface bands.

[0056] Figure 3 This is a schematic diagram of the segmented clipping region.

[0057] Figure 4 This is a schematic diagram showing the new isoparametric line being truncated by the clipped curve during reparameterization.

[0058] Figure 5 It is a schematic diagram of the new isoparametric lines of the surface in the model space.

[0059] Figure 6 It is a schematic diagram of the natural transition of the isoparametric line machining trajectory between two adjacent curved surfaces.

[0060] Figure 7 This is a schematic diagram of the bicycle seat cutting process; where a is the original model of the bicycle seat, b is the cutting curve of the top surface of the bicycle seat, c is the segmentation result of the cutting domain of the top surface of the bicycle seat, and d is the surface obtained by reparameterization and the generated processing trajectory. Detailed Implementation

[0061] The invention will now be further described with reference to the accompanying drawings.

[0062] like Figure 1 As shown, the implementation process and circumstances of this invention according to the complete method described above are as follows:

[0063] 1) After reconstructing the B-rep solid model composed of trimmed surfaces (i.e., removing the trimmed surfaces), a B-rep solid model composed of multiple complete tensor product surfaces within a given accuracy is obtained and denoted as the target B-rep solid model. In the target B-rep solid model, adjacent tensor product surfaces are watertightly connected, and two adjacent tensor product surfaces have the same node vector along the intersection direction (called isomorphism). The solid model is provided by the user, and the processing parameters are determined by the user. The functions of importing the model and inputting parameters are provided by a conventional CAD / CAM system. In this embodiment, the B-rep solid model of a bicycle seat is imported into the CAD / CAM system, such as... Figure 7 As shown in a.

[0064] 1) Specifically:

[0065] 1.1) Divide the clipping region of each clipped surface (i.e., the part of the surface retained after the intersection operation) into multiple curvilinear quadrilateral regions to obtain the initial surface corresponding to the current clipping region (e.g., ...). Figure 3 As shown), traverse all clipping surfaces in the B-rep solid model composed of clipping surfaces, divide to obtain the initial surfaces corresponding to all clipping surfaces, thereby obtaining the updated B-rep solid model; in this embodiment, the clipping curve of the bicycle seat top surface is as follows. Figure 7 As shown in b.

[0066] In section 1.1), the specific steps for processing the initial surface of each trimmed surface are as follows:

[0067] 1.1.1) Break all clipping curves within the current clipping domain at all non-differentiable points, so that all resulting sub-curve segments are smooth, thus obtaining multiple smooth curves;

[0068] 1.1.2) Along the feature points of the smooth curve, divide the entire smooth curve into corresponding sub-segments. Then, divide each sub-segment into a main u-feature segment and a main v-feature segment to obtain a set of feature segments. In specific implementation, points with slopes of 0, +1, -1, and infinity in the smooth curve are denoted as feature points. The sub-segment that mainly extends along the u direction is called the main u-feature segment, and the sub-segment that mainly extends along the v direction is called the main v-feature segment.

[0069] 1.1.3) Extract a sub-region from the current cropping domain based on the main extension direction of each feature segment in the feature segment set, such that each feature segment is a partial boundary of the sub-region. Traverse the feature segment set, taking all feature segments as partial boundaries of their corresponding sub-regions, thereby obtaining the preliminary segmentation result of the current cropping domain; for example... Figure 3 As shown, Figure 3 The parametric clipping curve is divided into a main u-segment and a main v-segment by a feature point, further dividing the parameter domain into three curvilinear quadrilateral regions: orange, green, and purple. The main u-segment is associated with the orange region and is bounded by a main u-segment curve, two straight lines parallel to the v-direction, and a boundary segment. Simply put, a main u-segment can be divided into a portion of the clipping domain by drawing straight lines along the v-direction at its two endpoints; this portion is the sub-region described above.

[0070] 1.1.4) Based on the preliminary segmentation result of the current clipping domain, the current clipping domain is further divided into several quadrilateral regions, and the topology information (topology information refers to the connection relationship between the newly generated quadrilateral regions) is updated to obtain the initial surface corresponding to the current clipping domain. In this embodiment, the clipping domain segmentation result of the top surface of the bicycle seat is as follows: Figure 7 As shown in c.

[0071] 1.2) Fit the new intersection lines between adjacent initial surfaces in the updated B-rep solid model, and then extract the isoparametric lines of each initial surface based on the new intersection lines;

[0072] 1.2) For each initial surface, the generation of its isoparametric lines specifically includes the following steps:

[0073] 1.2.1) After fitting scattered points at the non-watertight intersection of two adjacent initial surfaces, a definite curve is generated and used as the new intersection line. The non-watertightness is caused by two not entirely identical approximations of the intersection line by the parameter domains of the two surfaces. Points can be taken on these two approximation lines according to the user-defined precision, and then the two sets of points are fitted with a curve. Of course, the higher the precision required by the user, the more points are taken, the closer the newly fitted intersection line is to the intersection line of the original model, and the smaller the degree of deformation of the model reconstruction. The cost is the time and storage space consumed by the method.

[0074] 1.2.2) Extract multiple isoparametric lines from each trimmed surface and appropriately trim or extend these lines so that their endpoints fall on the new intersection line. Obtain the processed isoparametric lines and use them as the isoparametric lines of the corresponding initial surface in the parameter domain, such as... Figure 4 As shown;

[0075] Specifically, the truncation of isoparametric lines is accomplished by subdividing B-spline curves. Nodes are inserted at the intersection of the isoparametric lines and the contour, and then a segment within the clipping domain is taken.

[0076] The choice of isoparametric lines depends on the new intersection line fitted in 1.2.1: select the points corresponding to the Grievance abscissas of the fitted curve, project them onto the original surface, and then select the isoparametric line passing through the projection points. The parameters of the fitted curve C are different from the two parameters u and v of the original surface S; here, we call it C(ζ). Let C(ζ) be the point corresponding to each Grievance abscissa of C(ζ). i The projection onto the original surface is S(u) i ,v i If so, then we can choose v = v i These are isoparametric lines.

[0077] 1.2.3) After transforming the isoparametric lines of the initial surface in the parameter domain to the model space, the isoparametric lines of the initial surface in the model space are obtained, such as... Figure 5 As shown. Figure 5 Let 'a' represent the original isoparametric lines of the original surface. After intersecting with another surface, leaving an intersection line, the vertical isoparametric lines are truncated, causing their start and end parameter values ​​to differ. Therefore, reparameterization is used to give them the same start and end parameter values. Finally, the control points of the new surface are solved using these reparameterized isoparametric lines. The new surface is the lower half of the original surface, and its isoparametric lines are as follows: Figure 5 As shown in b. The function provided by the CAD / CAM library is used to determine the position of isoparametric lines in model space.

[0078] 1.3) By reparameterizing multiple isoparametric lines corresponding to each initial surface, the control points of each initial surface are obtained, and then each tensor product surface is obtained. Thus, a B-rep solid model composed of multiple complete tensor product surfaces is formed and denoted as the target B-rep solid model.

[0079] Because the coordinates of the control point grid of the B-spline surface and the coordinates of the control polygons of the B-spline isoparametric lines have a simple linear relationship, the control point grid of the newly generated surface can be obtained by solving a system of linear equations using a sufficient number of control polygons of isoparametric lines.

[0080] The original model's lack of watertightness stems from the fact that the approximation of the intersection line by the parameter domains of the two surfaces cannot be completely superimposed in the model space. However, after reparameterization, a suitable curve is found by fitting multiple scattered points near the original intersection point. The edge interpolation property of the B-spline surface is then used to ensure that both intersecting surfaces use this curve as their boundary. This achieves a watertight intersection of the two surfaces in the model space, guaranteeing the watertightness of the target B-rep solid model. The machining trajectory can continuously cross the intersection line without any "cracks" or "bulges" that affect the robustness of the method.

[0081] The node vectors of the new surface after reparameterization along the intersection direction are exactly the node vectors of C(ζ). Therefore, the node vectors of two adjacent surfaces are the same along the intersection direction, which is the homogeneity mentioned earlier.

[0082] In section 1.3), the reconstructed target B-rep entity model is composed of complete tensor product surface patches, and the surfaces are watertight. Furthermore, adjacent surfaces can be homogenized along the intersection direction, meaning the node vectors in the connection direction are identical. An isoparametric line can originate from one surface, pass through the watertight intersection line, and reach another surface, with its parameter values ​​remaining unchanged. The generation of the tensor product surface for each initial surface specifically includes the following steps:

[0083] 1.3.1) Perform a reparameterization transformation on the isoparameter lines corresponding to each initial surface in the model space so that the starting parameter value and the ending parameter value of each isoparameter line are the same, and obtain the transformed isoparameter lines;

[0084] 1.3.2) By utilizing the linear relationship between the control points of the B-spline surface and the control points of the transformed isoparametric lines, the control point mesh of the initial surface in the model space is obtained;

[0085] 1.3.3) Assign the control points of the new intersection line of the current initial surface to the control points on the edge of the control point grid of the initial surface in the model space, so that the B-spline surface in the control point grid is exactly interpolated at the new intersection line determined in 1.2.1, thus achieving water tightness;

[0086] 1.3.4) Use the node vectors and orders of the new intersection line as the node vectors and orders of the initial surface in the direction of the new intersection line, and use the node vectors and orders of the transformed isoparametric lines as the node vectors and orders of the initial surface in the direction of the isoparametric lines, thereby obtaining the tensor product surface corresponding to the current initial surface.

[0087] 2) Based on the topological connections between tensor product surfaces in the target B-rep solid model, the target B-rep solid model is divided into multiple tensor product surface bands. A tensor product surface band refers to a set of linearly arranged adjacent surfaces in the model. Except for the tensor product surface patches at both ends, each tensor product surface patch is adjacent to the two adjacent surface patches with a set of opposite edges, such as... Figure 2 As shown;

[0088] After model reconstruction, many adjacent surfaces are generated. Based on their topological connections, they are organized into multiple surface strips. In this way, each machining trajectory can sequentially pass through the surfaces on the surface strip. After each surface strip is generated, all its surfaces are removed from the model's patch set. Finally, a few surface patches may remain that cannot be grouped into strips; these isolated, non-trimmed surfaces are then machined individually using traditional methods.

[0089] Because isomorphism ensures that the node vectors of new adjacent surfaces are identical along the intersection direction, all surfaces on the surface strip have a single, identical node vector. The isoparametric machining path can then sequentially traverse each surface on the surface strip along its extension direction, maintaining unchanged parameter values, such as... Figure 6 As shown.

[0090] 2) Specifically:

[0091] 2.1) Construct two topological relationship tables based on the topological relationships of the target B-rep entity model. The two topological relationship tables are the adjacent surface table and the opposite edge table. When querying, the adjacent surface table can be used to find another surface that is adjacent to the edge of a surface based on a surface and one of its edges. When querying, the opposite edge table can be used to find the opposite edge of the edge of a surface based on a surface and one of its edges.

[0092] In step 1.1.4, when the original trimmed surface is segmented, the update of the model's topology is continuously recorded, so the complete topology information of the target B-rep solid model is available here.

[0093] 2.2) Based on the adjacent surface table and the opposite edge relation table, multiple tensor product surface bands of the target B-rep solid model are generated using the traversal method, such as... Figure 2 As shown, Figure 2 It is a model composed of 16 surfaces. The 4 gray surfaces form a surface band. When generating the surface band, first select surface A and add it to list L1. Then find B through edge e1, and find D through the opposite edge e3 of e1 on B. List L1 is {A,B,D}. A finds C through edge e2. The opposite edge e4 of e2 on C is the boundary of the model. Therefore, list L2 is {C}. After inverting L2 and splicing it with L1, the surface band {C,A,B,D} is formed.

[0094] 2.2) Specifically:

[0095] 2.2.1) The first set S is composed of all tensor product surfaces in the target B-rep entity model, and then two surface lists L1 and L2 are initialized with two empty lists;

[0096] 2.2.2) Select any tensor product surface A and a pair of its opposite edges e1 and e2 from the current first set S, and add the tensor product surface A to the first surface strip list L1;

[0097] 2.2.3) Obtain the tensor product surface B adjacent to edge e1 of tensor product surface A from the adjacent surface table, add tensor product surface B to the first surface patch list L1, and obtain the opposite edge e3 of edge e1 on tensor product surface B from the opposite edge relationship table. Then, find the adjacent surface of tensor product surface B with respect to edge e3 and add it to the first surface patch list L1. Continue until the edge of the target B-rep solid model is reached to obtain the final first surface patch list L1.

[0098] 2.2.4) If edge e2 is not at the boundary of the target B-rep solid model, find the tensor product surface C of the tensor product surface A with respect to edge e2 from the adjacent surface table, add the tensor product surface C to the second surface strip list L2, find the opposite edge e4 of edge e2 on the tensor product surface C from the opposite edge relationship table, find the tensor product surface C with respect to edge e4 and add it to the second surface strip list L2, until the edge of the target B-rep solid model is reached, and obtain the final second surface strip list L2;

[0099] 2.2.5) Reverse the final second surface strip list L2 and then concatenate it with the final first surface strip list L1 to generate a tensor product surface strip and save it. Then, delete all tensor product surfaces of the tensor product surface strip from the first set S and update the first set S.

[0100] 2.2.6) Repeat steps 2.2.1)-2.2.5) until the first set S becomes an empty set, and obtain multiple tensor product surface bands.

[0101] 3) Generate the isoparametric machining trajectories of all tensor product surfaces on each tensor product surface band to obtain the machining path of the entire model.

[0102] Because the machining process requires that the residual height not exceed the allowable value set by the user, and within a curved surface strip, the u-parameter value of a machining trajectory is the same for all surfaces on the strip (ensuring the continuity of the machining trajectory). Therefore, after using the isoparameteric line u=0 to pass through the surface strip, the u-parameter value of the next machining trajectory needs to be determined by all the surface patches in that surface strip, and cannot be determined solely by the shape of a single surface. In this embodiment, the surface obtained after reparameterizing the bicycle seat and the generated machining trajectory are as follows... Figure 7 As shown in d.

[0103] 3) Specifically:

[0104] 3.1) Use path planning methods to generate tool points on all tensor product surface bands;

[0105] 3.1.1) Specifically:

[0106] 3.1.1) Initialize the parameter values ​​in the direction of the new intersection line. In specific implementation, the parameter values ​​in the direction of the new intersection line are denoted as u parameter values, that is, u value is set to 0;

[0107] 3.1.2) Generate the tool position points on the isoparameter line corresponding to the parameter values ​​(u parameter values) in the direction of the new intersection line on the entire tensor product surface band according to the machining accuracy in the parameter domain;

[0108] 3.1.3) Move the tool position point in the direction of increasing parameter value in the new intersection line direction according to the set residual height allowable value. Then take the tool position point with the smallest increment in the new intersection line direction among the tool positions in the entire curved surface as the auto-increment value of parameter value in the new intersection line direction, and update the parameter value in the new intersection line direction in order to generate the next machining trajectory.

[0109] 3.1.4) Based on the value in the direction of the current new intersection line, repeat 3.1.2)-3.1.3) until all tool points of the entire tensor product surface band are generated, and obtain all tool points of the current tensor product surface band.

[0110] 3.2) Convert the tool position information into G-code instructions and use them for CNC machining.

[0111] This invention first accepts the solid model and machining path information processed in the above steps, and then displays them on the user interface for the user to perform further operations. The specific display technology is consistent with existing commercial CAD / CAM software.

[0112] The above specific embodiments are used to explain and illustrate the present invention, but not to limit the present invention. Any modifications and changes made to the present invention within the spirit and scope of the claims shall fall within the protection scope of the present invention.

Claims

1. A method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization, comprising the following steps: 1) After reconstructing the B-rep solid model composed of trimmed surfaces, a B-rep solid model composed of multiple complete tensor product surfaces is obtained and denoted as the target B-rep solid model. In the target B-rep solid model, adjacent tensor product surfaces are watertight connected, and two adjacent tensor product surfaces have the same node vector along the intersection direction. Specifically, 1) refers to: 1.1) Divide the clipping domain of each clipping surface into multiple curved quadrilateral regions to obtain the initial surface corresponding to the current clipping domain. Traverse all clipping surfaces in the B-rep solid model composed of clipping surfaces to obtain the initial surface corresponding to all clipping surfaces, thereby obtaining the updated B-rep solid model. 1.2) Fit the new intersection lines between adjacent initial surfaces in the updated B-rep solid model, and then extract the isoparametric lines of each initial surface based on the new intersection lines; 1.3) By reparameterizing the isoparameter lines of each initial surface, the control points of each initial surface are obtained, and then each tensor product surface is obtained. Thus, a B-rep solid model composed of multiple complete tensor product surfaces is formed and denoted as the target B-rep solid model. 2) Based on the topological connections between tensor product surfaces in the target B-rep solid model, divide the target B-rep solid model into multiple tensor product surface bands; 3) Generate the isoparametric machining trajectories of all tensor product surfaces on each tensor product surface band to obtain the machining path of the entire model.

2. The method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 1, characterized in that, In section 1.1), the specific steps for processing the initial surface of each trimmed surface are as follows: 1.1.1) Break all clipping curves within the current clipping domain at all non-differentiable points to obtain multiple smooth curves; 1.1.2) Along the feature points in the smooth curve, divide all the smooth curves into corresponding sub-segments, and then divide each sub-segment into the main u feature segment and the main v feature segment to obtain the feature sub-segment set; 1.1.3) Extract a sub-region in the current clipping domain based on the main extension direction of each feature segment in the feature segment set, so that each feature segment is a partial boundary of the sub-region. Traverse the feature segment set and use all feature segments as partial boundaries of the corresponding sub-regions to obtain the preliminary segmentation result of the current clipping domain. 1.1.4) Based on the preliminary segmentation results of the current clipping domain, the current clipping domain is further divided into several quadrilateral regions to obtain the initial surface corresponding to the current clipping domain.

3. The method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 1, characterized in that, As stated in 1.2), the generation of isoparametric lines for each initial surface specifically includes the following steps: 1.2.1) After fitting scattered points at the non-watertight intersection of two adjacent initial surfaces, a definite curve is generated and used as the new intersection line; 1.2.2) Extract multiple isoparametric lines from each trimmed surface and trim or extend multiple isoparametric lines so that the endpoints of multiple isoparametric lines fall on the current new intersection line, thereby obtaining multiple processed isoparametric lines and using them as the isoparametric lines of the corresponding initial surface in the parameter domain. 1.2.3) After transforming the isoparametric lines of the initial surface in the parameter domain to the model space, the isoparametric lines of the initial surface in the model space are obtained.

4. The method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 1, characterized in that, In section 1.3), the generation of the tensor product surface for each initial surface specifically includes the following steps: 1.3.1) Perform a reparameterization transformation on the isoparameter lines corresponding to each initial surface in the model space so that the starting parameter value and the ending parameter value of each isoparameter line are the same, and obtain the transformed isoparameter lines; 1.3.2) By utilizing the linear relationship between the control points of the B-spline surface and the control points of the transformed isoparametric lines, the control point mesh of the initial surface in the model space is obtained; 1.3.3) Assign the control points of the new intersection line of the current initial surface to the control points on the edge of the control point grid of the initial surface in the model space, so that the B-spline surface in the control point grid is interpolated at the new intersection line. 1.3.4) Use the node vectors and orders of the new intersection line as the node vectors and orders of the initial surface in the direction of the new intersection line, and use the node vectors and orders of the transformed isoparametric lines as the node vectors and orders of the initial surface in the direction of the isoparametric lines, thereby obtaining the tensor product surface corresponding to the current initial surface.

5. The method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 1, characterized in that, Specifically, 2) refers to: 2.1) Construct two topological relationship tables based on the topological relationships of the target B-rep entity model. The two topological relationship tables are the adjacent surface table and the opposite edge relationship table, respectively. 2.2) Based on the adjacent surface table and the opposite edge relation table, use the traversal method to generate multiple tensor product surface bands of the target B-rep solid model.

6. The method for generating parametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 5, characterized in that, Specifically, 2.2) refers to: 2.2.1) The first set S is composed of all tensor product surfaces in the target B-rep solid model, and then two surface lists L1 and L2 are initialized with two empty lists; 2.2.2) Select any tensor product surface A and a pair of its opposite edges e1 and e2 from the current first set S, and add the tensor product surface A to the first surface strip list L1; 2.2.3) Obtain the tensor product surface B adjacent to edge e1 of tensor product surface A from the adjacent surface table, add tensor product surface B to the first surface band list L1, and obtain the opposite edge e3 of edge e1 on tensor product surface B from the opposite edge relationship table. Then find the adjacent surface of tensor product surface B with respect to edge e3 and add it to the first surface band list L1. Continue until the edge of the target B-rep solid model is reached to obtain the final first surface band list L1. 2.2.4) If edge e2 is not at the boundary of the target B-rep solid model, find the tensor product surface C of the tensor product surface A with respect to edge e2 from the adjacent surface table, add the tensor product surface C to the second surface band list L2, find the opposite edge e4 of edge e2 on the tensor product surface C from the opposite edge relation table, find the tensor product surface C with respect to edge e4 and add it to the second surface band list L2, until the edge of the target B-rep solid model is reached, and obtain the final second surface band list L2; 2.2.5) Reverse the final second surface strip list L2 and concatenate it with the final first surface strip list L1 to generate a tensor product surface strip and save it. Then, delete all tensor product surfaces of the tensor product surface strip from the first set S and update the first set S. 2.2.6) Repeat 2.2.1)-2.2.5) until the first set S becomes an empty set, and obtain multiple tensor product surface bands.

7. The method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 1, characterized in that, Specifically, 3) refers to: 3.1) Use path planning methods to generate tool points on all tensor product surface bands; 3.2) Convert the tool position information into G-code instructions and use them for CNC machining.

8. The method for generating isoparametric processing paths for trimmed freeform surfaces based on reparameterization according to claim 7, characterized in that, Specifically, 3.1) refers to: 3.1.1) Initialize the parameter values ​​in the direction of the new intersection line; 3.1.2) Generate tool positions on the isoparameter lines corresponding to the parameter values ​​in the direction of the new intersection line on the entire tensor product surface band according to the machining accuracy in the parameter domain; 3.1.3) Move the tool position point in the direction of increasing parameter value in the new intersection line direction according to the set residual height allowable value. Then take the tool position point with the smallest increment in the new intersection line direction among the tool positions on the entire curved surface as the auto-increment value of parameter value in the new intersection line direction, and update the parameter value in the new intersection line direction. 3.1.4) Based on the parameter values ​​in the direction of the new intersection line, repeat steps 3.1.2)-3.1.3) until all tool points of the entire tensor product surface are generated, thus obtaining all tool points of the current tensor product surface.

Images