A method and system for automatic selection of impeller machining tools based on feature recognition

Abstract

The present application proposes a method and system for automatic selection of impeller machining tools based on feature recognition, which belongs to the field of aero-engine technology, wherein the method includes: using a feature recognition method of a graph to perform feature recognition on the three-dimensional model of the impeller, and then determining the diameter of the impeller machining tool; offsetting the surface of the impeller according to the current diameter of the impeller machining tool, and using a feature recognition method of a graph to perform feature recognition on the offset three-dimensional model of the impeller again, and dividing the blade feature points of the offset impeller into a blade root line point set and a blade surface point set; if all blade feature points in the blade root line point set are reachable, and all blade feature points in the surface point set are reachable, the maximum value of all shortest tool lengths is selected as the final shortest tool length. The present application can not only realize the automatic selection of impeller machining tools, but also reduce time complexity and speed up calculation speed.

Description

Technical Field

[0001] The present invention belongs to the technical field of aero-engines, and in particular relates to a method and system for automatically selecting impeller machining tools based on feature recognition. Background Art

[0002] The integral impeller is a key component in aircraft engines and a typically difficult-to-machine part. This is primarily due to the high interference structure and difficult-to-machine material of the integral impeller, making it difficult to achieve high machining efficiency. Choosing an appropriately sized machining tool without changing the material not only ensures machining feasibility but also effectively improves machining efficiency. Therefore, effectively and automatically selecting the appropriate machining tool size to improve machining efficiency and ensure product quality is crucial in integral impeller machining.

[0003] Generally speaking, tool dimensions primarily encompass diameter and length. Increasing tool diameter improves machining efficiency, but also increases the likelihood of interference. Tool length influences tool stiffness. Decreasing tool length increases stiffness, but also increases the likelihood of collision between the fixture and the workpiece.

[0004] To resolve this contradiction, several solutions exist in the prior art, such as the Chinese patent "CN116150994A: A Tool Selection Method for Cavity Feature Machining Based on Process Information Mining" and the reference "Tool Selection Method for Multi-Axis Machining of Complex Curved Channels," Journal of Mechanical Engineering, Vol. 54, No. 3, 117-124. However, these methods result in the maximum tool diameter obtained from a large number of feature points being larger than the final tool diameter, resulting in a significant amount of redundant calculations and significantly increasing the algorithm's time complexity. Summary of the Invention

[0005] In response to the deficiencies of the prior art, this application proposes a method and system for automatically selecting impeller machining tools based on feature recognition.

[0006] In a first aspect, the present application proposes a method for automatically selecting impeller machining tools based on feature recognition, comprising:

[0007] Step S1: Obtaining a three-dimensional model of an impeller;

[0008] Step S2: Using a graph feature recognition method to perform feature recognition on the impeller three-dimensional model to obtain the impeller blade feature points;

[0009] Step S3: obtaining the shortest distance between blades and the minimum curvature radius on the blades according to the blade feature points of the impeller;

[0010] Step S4: determining the diameter of the impeller machining tool according to the shortest distance between the blades and the minimum curvature radius on the blades;

[0011] Step S5: offsetting the surface of the impeller according to the current diameter of the impeller machining tool, and performing feature recognition again on the offset impeller three-dimensional model using a graph feature recognition method to obtain blade feature points of the offset impeller;

[0012] Step S6: dividing the offset impeller blade feature points into a blade root line point set and a blade surface point set;

[0013] Step S7: If all blade feature points in the blade root line point set are reachable, go to step S8; otherwise, subtract the preset offset from the current impeller machining tool diameter, return to step S5, and perform the offset again;

[0014] Step S8: If all blade feature points in the blade surface point set are reachable, compare the shortest tool lengths obtained from the blade root line point set and the blade surface point set, select the maximum value of all the shortest tool lengths as the final shortest tool length, and use the final shortest tool length and the current impeller machining tool diameter as the result of automatic selection of the impeller machining tool. Otherwise, subtract the preset offset from the current impeller machining tool diameter, return to step S5, and re-offset.

[0015] The feature recognition method using a graph is used to perform feature recognition on the impeller three-dimensional model to obtain the impeller blade feature points, including:

[0016] Step S2.1: Obtain the topological tree of the impeller 3D model, the attributes of all faces, and the concavity and convexity of the connecting edges;

[0017] Step S2.2: Obtain the adjacent relationship of the faces in the impeller three-dimensional model according to the topological tree, and construct the initial adjacency matrix A, where A ij is the adjacency attribute between the i-th face and the j-th face, A ji is the adjacency attribute between the jth face and the ith face;

[0018] Step S2.3: Traverse the initial adjacency matrix A. If face i is connected to face j, let A ij =A ji =1, if face i and face j are not connected, then let A ij =A ji =0, and get the adjacency matrix A after traversal;

[0019] Step S2.4: Based on the attributes of the face and the concavity of the connecting edge, the traversed adjacency matrix A is updated to the attribute adjacency matrix B, where B ij is the concavity of the connecting edge between the i-th and j-th surfaces, Bii is the attribute of the i-th face;

[0020] Step S2.5: Traverse the attribute adjacency matrix B. If the connected faces i and j need to be split, let B ij =B ji =0, and obtain the attribute adjacency matrix B after segmentation;

[0021] Step S2.6: Use a depth-first traversal algorithm to extract a submatrix from the segmented attribute adjacency matrix B, and perform encoding to compress the multi-dimensional information into one-dimensional information to obtain an encoded submatrix;

[0022] Step S2.7: Compare the encoded sub-matrix with the predefined feature library one by one. If the encoded sub-matrix is ​​consistent with the features in the predefined feature library, the impeller blade feature points corresponding to the encoded sub-matrix are obtained.

[0023] Determining the diameter of the impeller machining tool based on the shortest spacing between the blades and the minimum curvature radius on the blades includes: selecting the minimum value between the shortest spacing between the blades and the minimum curvature radius on the blades as the diameter of the impeller machining tool.

[0024] The reachability includes: simplifying the reachable direction cone on the Gaussian sphere by using the proportion of the reachable direction cone in each spherical zone to obtain the simplified reachability.

[0025] The simplified reachability is calculated as follows:

[0026]

[0027] in, i is the serial number of the current ball zone, C i is the proportion of the reachable direction cone in each spherical zone of the Gaussian sphere, T is the step distance, R is the diameter of the current impeller machining tool, θ i is the angle between the i-th ray and the Z axis on the surface YOZ, wherein a three-dimensional Cartesian coordinate system is established in the hemisphere space of the Gaussian sphere with the center of the Gaussian sphere as the origin O, and the surface YOZ is a plane in the three-dimensional Cartesian coordinate system.

[0028] The shortest tool length is calculated as follows:

[0029]

[0030] Among them, L is the shortest tool length, R is the diameter of the current impeller machining tool, (x p ,y p ,z p ) is the coordinate of the impeller blade feature point after offset, (x p′ ,y p′,z p′ ) is the coordinate of the intersection of the tool axis vector and the outer surface of the blank.

[0031] In a second aspect, the present application proposes an automatic impeller machining tool selection system based on feature recognition, comprising:

[0032] A model acquisition module is used to obtain a three-dimensional model of the impeller;

[0033] The first feature recognition module is used to perform feature recognition on the impeller three-dimensional model using a graph feature recognition method to obtain the impeller blade feature points;

[0034] A data calculation module is used to obtain the shortest spacing between blades and the minimum curvature radius on the blades according to the blade feature points of the impeller;

[0035] a diameter determination module, configured to determine the diameter of an impeller machining tool according to the shortest distance between the blades and the minimum curvature radius on the blades;

[0036] The second feature recognition module is used to offset the surface of the impeller according to the diameter of the current impeller processing tool, and use the feature recognition method of the graph to perform feature recognition on the offset impeller three-dimensional model again to obtain the blade feature points of the offset impeller;

[0037] A point set division module, configured to divide the offset impeller blade feature points into a blade root line point set and a blade surface point set;

[0038] The first reachability judgment module is used to switch to the second reachability judgment module if all blade feature points in the blade root line point set are reachable. Otherwise, the current impeller machining tool diameter is subtracted by a preset offset, and the module returns to the second feature recognition module to re-offset.

[0039] The second reachability judgment module is used to compare the shortest tool lengths obtained from the blade root line point set and the blade surface point set if all blade feature points in the blade surface point set are reachable, select the maximum value of all the shortest tool lengths as the final shortest tool length, and use the final shortest tool length and the current impeller machining tool diameter as the result of automatic selection of the impeller machining tool. Otherwise, the current impeller machining tool diameter is subtracted by a preset offset, and the module is returned to the second feature recognition module for re-offsetting.

[0040] In a third aspect, the present application proposes an electronic device comprising: one or more processors, and a memory, wherein the memory is used to store instructions, and when the instructions are executed by the one or more processors, the one or more processors execute the method for automatic selection of impeller machining tools based on feature recognition.

[0041] In a fourth aspect, the present application proposes a computer-readable storage medium storing executable instructions, which, when executed, enable a processor to execute the method for automatic selection of impeller machining tools based on feature recognition.

[0042] In a fifth aspect, the present application proposes a computer program product, comprising a computer program or instructions, which, when executed by a processor, implements the method for automatic selection of impeller machining tools based on feature recognition.

[0043] Beneficial effects:

[0044] This application proposes a method and system for automatically selecting impeller machining tools based on feature recognition. This system introduces feature recognition technology, eliminating the need for manual selection of target surfaces, interference surfaces, and maximum selectable diameters. Furthermore, based on the feature recognition results, the system divides discrete feature point sets into blade root line point sets and blade surface point sets. This prioritizes blade root line point sets, which generally have poor accessibility, reducing time complexity and accelerating calculations. Furthermore, the system considers the maximum diameter and minimum length of the tool, and can adjust the minimum accessible direction cone ratio, enabling more detailed adjustments to the accessibility of the machining tool size. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 Flowchart of the method for automatically selecting impeller machining tools based on feature recognition according to an embodiment of the present application;

[0046] Figure 2 A flow chart of a method for automatically selecting impeller machining tools based on feature recognition according to an embodiment of the present application;

[0047] Figure 3 Schematic diagram of the feature recognition principle of the embodiment of the present application;

[0048] Figure 4 Schematic diagram of discrete blade surfaces according to an embodiment of the present application;

[0049] Figure 5 Simplified direction cone ratio distribution diagram of the embodiment of the present application;

[0050] Figure 6 The shortest tool length distribution diagram of the embodiment of the present application;

[0051] Figure 7 Schematic diagram of the actual reachable direction cone and the simplified reachable direction cone in the embodiment of the present application;

[0052] Figure 8 Schematic diagram of step length and relative error according to an embodiment of the present application;

[0053] Figure 9 Schematic diagram of error calculation evaluation value in an embodiment of the present application;

[0054] Figure 10 Schematic diagram of the ball zone according to the embodiment of the present application;

[0055] Figure 11 Schematic diagram of the theoretical shortest tool length according to an embodiment of the present application;

[0056] Figure 12 The embodiment of the present application provides an automatic selection system for impeller machining tools based on feature recognition. DETAILED DESCRIPTION

[0057] The specific implementation of the present application is further described in detail below with reference to the accompanying drawings and examples.

[0058] Example 1:

[0059] This embodiment proposes an automatic selection method for impeller machining tools based on feature recognition, such as Figure 1 、 Figure 2 Shown, including:

[0060] Step S1: Obtaining a three-dimensional model of an impeller;

[0061] Step S2: Using a graph feature recognition method to perform feature recognition on the impeller 3D model to obtain the impeller blade feature points, including:

[0062] Step S2.1: Obtain the topological tree of the impeller 3D model, the attributes of all faces, and the concavity and convexity of the connecting edges;

[0063] Step S2.2: Obtain the adjacent relationship of the faces in the impeller three-dimensional model according to the topological tree, and construct the initial adjacency matrix A, where A ij is the adjacency attribute between the i-th face and the j-th face, A ji is the adjacency attribute between the jth face and the ith face;

[0064] Step S2.3: Traverse the initial adjacency matrix A. If face i is connected to face j, let A ij =A ji =1, if face i and face j are not connected, then let A ij =A ji =0, and get the adjacency matrix A after traversal;

[0065] Step S2.4: Based on the attributes of the face and the concavity of the connecting edge, the traversed adjacency matrix A is updated to the attribute adjacency matrix B, where B ij is the concavity of the connecting edge between the i-th and j-th surfaces, B ii is the attribute of the i-th face;

[0066] Step S2.5: Traverse the attribute adjacency matrix B. If the connected faces i and j need to be split, let B ij =B ji=0, and obtain the attribute adjacency matrix B after segmentation;

[0067] Step S2.6: Use a depth-first traversal algorithm to extract a submatrix from the segmented attribute adjacency matrix B, and perform encoding to compress the multi-dimensional information into one-dimensional information to obtain an encoded submatrix;

[0068] Step S2.7: Compare the encoded sub-matrix with the predefined feature library one by one. If the encoded sub-matrix is ​​consistent with the features in the predefined feature library, the impeller blade feature points corresponding to the encoded sub-matrix are obtained.

[0069] In this embodiment, the image feature recognition algorithm includes three main steps: extracting feature information, constructing and decomposing an attribute adjacency matrix, and identifying features and their sub-features. During the feature information extraction phase, attention is paid to the model's topology, geometry, and concave-convex edge information. Topological information describes the hierarchical relationships between volumes, faces, edges, and points; geometric information indicates face types; and concave-convex edge information illustrates the connections between faces, providing the basis for the subsequent decomposition of the attribute adjacency matrix. The attribute adjacency matrix is ​​another representation of feature information. It stores the acquired feature information in a matrix and then splits the attribute adjacency matrix based on the structural characteristics of the impeller. After decomposition, the original attribute adjacency matrix is ​​divided into multiple sub-matrices, each corresponding to a feature. After feature recognition, some sub-matrices are further used to identify sub-features. For example, the sub-matrix of the blade feature is subdivided into blade wall, blade fillet, and blade cover. Feature and sub-feature recognition relies on a predefined feature library. Only features and sub-features already in the predefined feature library are recognized. Accordingly, adding new features to the predefined feature library also increases the number of identifiable features.

[0070] Step S3: obtaining the shortest distance between blades and the minimum curvature radius on the blades according to the blade feature points of the impeller;

[0071] Step S4: determining the diameter of the impeller machining tool according to the shortest spacing between the blades and the minimum curvature radius on the blades, including: selecting the minimum value between the shortest spacing between the blades and the minimum curvature radius on the blades as the diameter of the impeller machining tool.

[0072] In this embodiment, a method for automatically selecting impeller machining tools based on feature recognition includes three major steps. Step 1 includes: Steps S1 to S4. First, the impeller model is imported into UG software. The impeller is abstracted into an attribute adjacency graph using a graph-based feature recognition method. Then, the impeller is segmented according to its structural characteristics and compared with a predefined feature library to identify the blade features of the impeller, such as Figure 3Finally, based on this, the shortest distance H between the blades and the minimum curvature radius ρ on the blade are obtained, the two are compared, and the minimum value of the two quality inspections is taken to determine the initial diameter d0 of the tool.

[0073] Step S5: offsetting the surface of the impeller according to the current diameter of the impeller machining tool, and performing feature recognition again on the offset impeller three-dimensional model using a graph feature recognition method to obtain blade feature points of the offset impeller;

[0074] Step S6: dividing the offset impeller blade feature points into a blade root line point set and a blade surface point set;

[0075] Step S7: If all blade feature points in the blade root line point set are reachable, go to step S8; otherwise, subtract the preset offset from the current impeller machining tool diameter, return to step S5, and perform the offset again;

[0076] Step S8: If all blade feature points in the blade surface point set are reachable, compare the shortest tool lengths obtained from the blade root line point set and the blade surface point set, select the maximum value of all the shortest tool lengths as the final shortest tool length, and use the final shortest tool length and the current impeller machining tool diameter as the result of automatic selection of the impeller machining tool. Otherwise, subtract the preset offset from the current impeller machining tool diameter, return to step S5, and re-offset.

[0077] In this embodiment, step 2 includes: steps S5 to S8, offsetting the impeller surface according to the current tool diameter, and then using the feature recognition method again to identify the offset blade features, and discretizing the blade into several feature points along the U and V directions, such as Figure 4 And according to the distribution diagram of the simplified reachable direction cone ratio and the theoretical shortest tool length on the blade, it can be found that the maximum tool diameter and the shortest tool length both appear at the blade root, as shown in Figure 5 、 Figure 6 As shown. Therefore, the feature point set is further divided into the blade root line point set and the blade surface point set. When judging the feasibility of the tool diameter, the blade root line point set is traversed first. At each feature point, a number of tool axis vectors are emitted with a certain step distance, and the interference between these tool axis vectors and the offset surface is judged. Then, the reachable direction cone is obtained based on the tool axis vector without interference, and the reachability of this feature point is judged. However, considering that the actual size of the reachable direction cone is not easy to obtain accurately, it is simplified to a simplified reachable direction cone, as shown in Figure 7 shown.

[0078] In the absence of any interference, the reachable direction cone of the tool axis vector will map a complete sphere on the Gaussian sphere. However, in reality, the tool axis vector is limited by the range of the machine tool. The actual reachable range is the upper hemisphere of the original sphere. Considering the interference of the workpiece, the actual reachable direction cone can only map a part of the sphere on the Gaussian sphere, such as Figure 7 The area enclosed by the irregular curve in the figure is shown. However, impeller shapes are generally complex, and the actual reachable direction cone obtained is often irregular, making it difficult to accurately determine its size. Therefore, some simplification is necessary to quickly approximate the reachable direction cone. Based on this, we further calculate the proportion of the reachable direction cone in the overall reachable direction cone to make the reachability more intuitive. For ease of presentation, this will be referred to as the reachable direction cone proportion.

[0079] Several rays emitted from the center of the sphere will interfere with the workpiece. After traversing all tool axis vectors, the boundary of the non-interference tool axis vector set is obtained. Then, a simplified reachable direction cone can be constructed through the boundary of the non-interference tool axis vector set, such as Figure 7 As shown on the right. There will inevitably be an error between the simplified feasible direction cone and the actual feasible direction cone, but the error value will decrease as the step size decreases. Therefore, it is necessary to study the relationship between the error value and the step size and select an appropriate step size. Since the focus in this article is on the minimum value of the achievable direction cone, the size of the minimum value determines whether the tool can reach the blade surface smoothly. Therefore, the point with the smallest achievable direction cone is specially selected as the research object. After selecting the ray source point, different step sizes are set in turn. The proportion of the achievable direction cone and the relative error obtained by each step size can be seen. Figure 8 In the test, a total of 7 different step lengths were set from small to large. The proportion of the reachable direction cone in the hemisphere space increased with the increase of the step length and began to converge when the step length was 0.5°. Therefore, it is recommended to approximate the result when the step length is 0.3° as the actual reachable direction cone proportion of the current point, and use this as a standard to obtain the relative error of other step lengths. The specific results are as follows: Figure 8 As shown by the broken line. Figure 8 The data in show that reducing the step size can indeed reduce the error between the simplified direction cone and the actual direction cone, but at the same time it will cause the calculation to increase rapidly. Therefore, it is necessary to minimize the amount of calculation while maintaining a small error. To this end, the following error calculation evaluation formula is proposed:

[0080]

[0081] Among them, T is the step size, S T is the proportion of the reachable direction cone with step size T. The above formula consists of two parts: the relative error of the reachable direction cone proportion and the inverse of the order of magnitude of the computational amount. Considering both the relative error and the computational amount, the smaller the result, the more appropriate the current step size. Figure 8 Evaluate each step length and obtain Figure 9 .according to Figure 9 Based on the evaluation value in , T = 5° step length is selected as the step length for obtaining the simplified reachable direction cone.

[0082] The simplified expression of the reachable direction cone is as follows:

[0083]

[0084] in, i is the serial number of the current ball zone, such as Figure 10 As shown. C i is the proportion of the reachable direction cone in each spherical zone of the Gaussian sphere, T is the step distance, R is the diameter of the current impeller machining tool, θ i is the angle between the i-th ray and the Z axis on the surface YOZ, θ is the angle between the Z axis on the surface YOZ of a number of rays emitted from the center of the Gaussian sphere with a step distance of T in the hemisphere space of the Gaussian sphere, wherein a three-dimensional Cartesian coordinate system is established in the hemisphere space of the Gaussian sphere with the center of the Gaussian sphere as the origin O, and the surface YOZ is a plane in the three-dimensional Cartesian coordinate system.

[0085] Count the sizes of the reachable direction cones of all feature points in the leaf root line point set. If the minimum value reaches the predetermined lower limit, perform the same reachability analysis on the feature points in the leaf surface point set. If it is not feasible, reduce the tool diameter and repeat step 2. Count the sizes of the reachable direction cones of all feature points. If the minimum value reaches the predetermined lower limit, proceed to the next step. If it is not feasible, reduce the tool diameter and repeat step 2. Dividing the point set into the leaf root line point set and the leaf surface point set helps to reduce the time complexity. If the blade wall is discretized into m scattered points, of which the number of scattered points of the leaf root line is k, the actual number of offsets is n, and the number of times the leaf root line point set is satisfied is i, then the time complexity before segmentation is as follows:

[0086]

[0087] The time complexity after segmentation is as follows:

[0088]

[0089] Among them, p is the p-th scatter point, and q is the q-th bias.

[0090] By comparing the time complexity expressions before and after segmentation, it can be found that segmenting the point set (i.e. dividing the set into leaf root line point set and leaf surface point set) can effectively reduce the time complexity.

[0091] In this embodiment, step three includes the shortest tool length obtained by comparing the blade root line point set and the blade surface point set in step 8. Based on step two, step three traverses all feature points and obtains the theoretical shortest tool length of a feature point based on the intersection of its non-interference tool axis vector and the blank surface. The calculation formula is as follows:

[0092]

[0093] Among them, L is the shortest tool length, R is the diameter of the current impeller machining tool, (x p ,y p ,z p ) is the coordinate of the impeller blade feature point after offset, (x p′ ,y p′ ,z p′ ) is the coordinate of the intersection of the tool axis vector and the outer surface of the blank, such as Figure 11 Then, the longest theoretical shortest tool length is found among all the feature points and used as the theoretical shortest tool length of the tool. At this point, the maximum diameter and shortest length of the tool are obtained.

[0094] This application proposes a method for automatic tool selection for impeller machining based on feature recognition. By introducing feature recognition technology, the method automatically identifies the various features of the impeller, eliminating the need for manual selection of target and interference surfaces. The method also obtains the maximum selectable tool diameter while acquiring the target surface. To further reduce the algorithm's time complexity, the discrete feature point set is divided into a blade root line point set and a blade surface point set, combining the feature recognition results. Prioritizing the blade root line point set, which generally has poor accessibility, reduces time complexity.

[0095] Example 2:

[0096] This embodiment proposes an automatic selection system for impeller machining tools based on feature recognition, such as Figure 12 Shown, including:

[0097] A model acquisition module is used to obtain a three-dimensional model of the impeller;

[0098] The first feature recognition module is used to perform feature recognition on the impeller three-dimensional model using a graph feature recognition method to obtain the impeller blade feature points;

[0099] A data calculation module is used to obtain the shortest spacing between blades and the minimum curvature radius on the blades according to the blade feature points of the impeller;

[0100] a diameter determination module, configured to determine the diameter of an impeller machining tool according to the shortest distance between the blades and the minimum curvature radius on the blades;

[0101] The second feature recognition module is used to offset the surface of the impeller according to the diameter of the current impeller processing tool, and use the feature recognition method of the graph to perform feature recognition on the offset impeller three-dimensional model again to obtain the blade feature points of the offset impeller;

[0102] A point set division module, configured to divide the offset impeller blade feature points into a blade root line point set and a blade surface point set;

[0103] The first reachability judgment module is used to switch to the second reachability judgment module if all blade feature points in the blade root line point set are reachable. Otherwise, the current impeller machining tool diameter is subtracted by a preset offset, and the module returns to the second feature recognition module to re-offset.

[0104] The second reachability judgment module is used to compare the shortest tool lengths obtained from the blade root line point set and the blade surface point set if all blade feature points in the blade surface point set are reachable, select the maximum value of all the shortest tool lengths as the final shortest tool length, and use the final shortest tool length and the current impeller machining tool diameter as the result of automatic selection of the impeller machining tool. Otherwise, the current impeller machining tool diameter is subtracted by a preset offset, and the module is returned to the second feature recognition module for re-offsetting.

[0105] Example 3:

[0106] This embodiment proposes an electronic device, comprising: one or more processors, and a memory, wherein the memory is used to store instructions, and when the instructions are executed by the one or more processors, the one or more processors execute the automatic selection method for impeller machining tools based on feature recognition.

[0107] The electronic device can be a mobile phone, computer, or tablet computer, and includes a memory and a processor. The memory stores a computer program that, when executed by the processor, implements the method for automatically selecting an impeller machining tool based on feature recognition as described in the embodiments. It is understood that the electronic device may also include an input / output (I / O) interface and a communication component.

[0108] The processor is configured to execute all or part of the steps of the method for automatically selecting an impeller machining tool based on feature recognition as described in the above embodiment. The memory is configured to store various types of data, such as instructions for any application or method in the electronic device, as well as data related to the application.

[0109] The processor can be an application specific integrated circuit (ASIC), a digital signal processor (DSP), a programmable logic device (PLD), a field programmable gate array (FPGA), a controller, a microcontroller, a microprocessor or other electronic components, and is used to execute the automatic selection method of impeller machining tools based on feature recognition described in the above embodiment.

[0110] Example 4:

[0111] This embodiment provides a computer-readable storage medium storing executable instructions. When the instructions are executed, if they are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium.

[0112] The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, server, or network device, etc.) to execute all or part of the steps of the automatic selection method for impeller machining tools based on feature recognition described in various embodiments of the present application.

[0113] The aforementioned storage media include: flash memory, hard disk, multimedia card, card-type memory (for example, SD (Secure Digital Memory Card) or DX (Memory Data Register, MDR abbreviation, memory data register) memory, etc.), random access memory (RAM), static random access memory (SRAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), programmable read-only memory (PROM), magnetic memory, disk, CD, server, APP (Application, abbreviation of application software) application store and other media that can store program verification codes, on which a computer program is stored. When the computer program is executed by the processor, the various steps of the above-mentioned method for automatic selection of impeller processing tools based on feature recognition can be implemented.

[0114] Example 5:

[0115] This embodiment provides a computer program product, including a computer program or instructions, which, when executed by a processor, implements the method for automatically selecting an impeller machining tool based on feature recognition.

[0116] Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution can be embodied in the form of a computer program product.

[0117] The various embodiments in this application are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0118] The scope of protection of this application is not limited to the above-described embodiments. Obviously, those skilled in the art may make various modifications and variations to this disclosure without departing from the scope and spirit of this disclosure. If such modifications and variations fall within the scope of the claims of this disclosure and their equivalents, the disclosure is intended to include such modifications and variations.

Claims

1. A method for automatically selecting impeller machining tools based on feature recognition, characterized in that: include: Step S1: Obtaining a three-dimensional model of an impeller; Step S2: Using a graph feature recognition method to perform feature recognition on the impeller three-dimensional model to obtain the impeller blade feature points; Step S3: obtaining the shortest distance between blades and the minimum curvature radius on the blades according to the blade feature points of the impeller; Step S4: determining the diameter of the impeller machining tool according to the shortest distance between the blades and the minimum curvature radius on the blades; Step S5: offsetting the surface of the impeller according to the current diameter of the impeller machining tool, and performing feature recognition again on the offset impeller three-dimensional model using a graph feature recognition method to obtain blade feature points of the offset impeller; Step S6: dividing the offset impeller blade feature points into a blade root line point set and a blade surface point set; Step S7: If all blade feature points in the blade root line point set are reachable, go to step S8; otherwise, subtract the preset offset from the current impeller machining tool diameter, return to step S5, and perform the offset again; Step S8: If all blade feature points in the blade surface point set are reachable, compare the shortest tool lengths obtained from the blade root line point set and the blade surface point set, select the maximum value of all the shortest tool lengths as the final shortest tool length, and use the final shortest tool length and the current impeller machining tool diameter as the result of automatic selection of the impeller machining tool. Otherwise, subtract the preset offset amount from the current impeller machining tool diameter, return to step S5, and re-offset.

2. The method for automatically selecting impeller machining tools based on feature recognition according to claim 1, characterized in that: The feature recognition method using a graph is used to perform feature recognition on the impeller three-dimensional model to obtain the impeller blade feature points, including: Step S2.1: Obtain the topological tree of the impeller 3D model, the attributes of all faces, and the concavity and convexity of the connecting edges; Step S2.2: Obtain the adjacent relationship of the faces in the impeller three-dimensional model according to the topological tree, and construct the initial adjacency matrix A, where A ij is the adjacency attribute between the i-th face and the j-th face, A ji is the adjacency attribute between the jth face and the ith face; Step S2.3: Traverse the initial adjacency matrix A. If face i is connected to face j, let A ij =A ji =1, if face i and face j are not connected, then let A ij =A ji =0, and get the adjacency matrix A after traversal; Step S2.4: Based on the attributes of the face and the concavity of the connecting edge, the traversed adjacency matrix A is updated to the attribute adjacency matrix B, where B ij is the concavity of the connecting edge between the i-th and j-th surfaces, B ii is the attribute of the i-th face; Step S2.5: Traverse the attribute adjacency matrix B. If the connected faces i and j need to be split, let B ij =B ji =0, and obtain the attribute adjacency matrix B after segmentation; Step S2.6: Use a depth-first traversal algorithm to extract a submatrix from the segmented attribute adjacency matrix B, and perform encoding to compress the multi-dimensional information into one-dimensional information to obtain an encoded submatrix; Step S2.7: Compare the encoded sub-matrix with the predefined feature library one by one. If the encoded sub-matrix is ​​consistent with the features in the predefined feature library, the impeller blade feature points corresponding to the encoded sub-matrix are obtained.

3. The automatic selection method for impeller machining tools based on feature recognition according to claim 1, characterized in that: Determining the diameter of the impeller machining tool based on the shortest spacing between the blades and the minimum curvature radius on the blades includes: selecting the minimum value between the shortest spacing between the blades and the minimum curvature radius on the blades as the diameter of the impeller machining tool.

4. The automatic selection method for impeller machining tools based on feature recognition according to claim 1, characterized in that: The reachability includes: simplifying the reachable direction cone on the Gaussian sphere by using the proportion of the reachable direction cone in each spherical zone to obtain the simplified reachability.

5. The method for automatically selecting impeller machining tools based on feature recognition according to claim 4, characterized in that: The simplified reachability is calculated as follows: in, i is the serial number of the current ball zone, C i is the proportion of the reachable direction cone in each spherical zone of the Gaussian sphere, T is the step distance, R is the diameter of the current impeller machining tool, θ i is the angle between the i-th ray and the Z axis on the surface YOZ, wherein a three-dimensional Cartesian coordinate system is established in the hemisphere space of the Gaussian sphere with the center of the Gaussian sphere as the origin O, and the surface YOZ is a plane in the three-dimensional Cartesian coordinate system.

6. The method for automatically selecting impeller machining tools based on feature recognition according to claim 1, characterized in that: The shortest tool length is calculated as follows: Among them, L is the shortest tool length, R is the diameter of the current impeller machining tool, (x p ,y p ,z p ) is the coordinate of the impeller blade feature point after offset, (x p′ ,y p′ ,z p′ ) is the coordinate of the intersection of the tool axis vector and the outer surface of the blank.

7. An automatic impeller machining tool selection system based on feature recognition, characterized in that: include: A model acquisition module is used to obtain a three-dimensional model of the impeller; The first feature recognition module is used to perform feature recognition on the impeller three-dimensional model using a graph feature recognition method to obtain the impeller blade feature points; A data calculation module is used to obtain the shortest spacing between blades and the minimum curvature radius on the blades according to the blade feature points of the impeller; a diameter determination module, configured to determine the diameter of an impeller machining tool according to the shortest distance between the blades and the minimum curvature radius on the blades; The second feature recognition module is used to offset the surface of the impeller according to the diameter of the current impeller processing tool, and use the feature recognition method of the graph to perform feature recognition on the offset impeller three-dimensional model again to obtain the blade feature points of the offset impeller; A point set division module, configured to divide the offset impeller blade feature points into a blade root line point set and a blade surface point set; The first reachability judgment module is used to switch to the second reachability judgment module if all blade feature points in the blade root line point set are reachable. Otherwise, the current impeller machining tool diameter is subtracted by a preset offset, and the module returns to the second feature recognition module to re-offset. The second reachability judgment module is used to compare the shortest tool lengths obtained from the blade root line point set and the blade surface point set if all blade feature points in the blade surface point set are reachable, select the maximum value of all the shortest tool lengths as the final shortest tool length, and use the final shortest tool length and the current impeller machining tool diameter as the result of automatic selection of the impeller machining tool. Otherwise, the current impeller machining tool diameter is subtracted from a preset offset, and the result is returned to the second feature recognition module for re-offsetting.

8. An electronic device, characterized in that: include: One or more processors, and a memory, wherein the memory is used to store instructions, and when the instructions are executed by the one or more processors, the one or more processors execute the method for automatically selecting an impeller machining tool based on feature recognition as described in any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that The device stores executable instructions, which, when executed, enable a processor to execute the method for automatically selecting an impeller machining tool based on feature recognition as claimed in any one of claims 1 to 6.

10. A computer program product comprising a computer program or instructions, characterized in that When the computer program or instruction is executed by a processor, the method for automatically selecting an impeller machining tool based on feature recognition as described in any one of claims 1 to 6 is implemented.

Images