Part processing intelligent process design method and system

Abstract

The present application relates to a part processing intelligent process design method and system, belonging to the mechanical engineering technical field, first pretreats the MBD three-dimensional model of the part, obtains the feature view, PMI information and process picture; the processing feature recognizer based on deep neural network YOLOv5 is used to identify processing feature; according to PMI information extraction method, the process information on processing feature is extracted;Through the comprehensive weighting algorithm, the processing feature processing element is matched in the process knowledge base;Processing element is sorted by using processing element intelligent sorting and optimization algorithm, and the optimal process route is generated;Through the man-machine interaction mode, other basic information of process is collected, and process card is generated.The present application combines part processing feature recognition and intelligent process design, makes design information and manufacturing information intercommunication, realizes the purpose of effectively using three-dimensional design model for process design, and can quickly obtain efficient and reasonable processing process route, improves the intelligentization of process design.

Description

Technical Field

[0001] This invention relates to the field of mechanical engineering technology, and in particular to an intelligent process design method and system for parts processing. Background Technology

[0002] The deep integration of digital technology and manufacturing technology is triggering far-reaching industrial changes, forming new manufacturing models, production methods, and industrial forms.

[0003] Process design is one of the main tasks of the technical departments in manufacturing enterprises. Its quality and efficiency have a significant impact on production organization, product quality, product cost, productivity, and production cycle. With the development of computer technology, computer-aided process design systems have been widely used in process design. The advantages of using such systems include: process engineers can quickly compile complete and detailed process plans and documents, greatly improving work efficiency, shortening process preparation time, and accelerating the production of new products. Furthermore, automatic feature recognition is a core technology and key support in the field of intelligent design and manufacturing. In all computer-aided design and manufacturing systems that rely on feature analysis and decision-making, feature recognition is an indispensable basic component.

[0004] However, current process design systems still have problems such as insufficient intelligence, digitalization, and automation. In particular, existing process design systems cannot effectively utilize three-dimensional design models for process design. Summary of the Invention

[0005] To address the problems existing in current process design systems, this invention proposes an intelligent process design method and system for parts machining. This method and system intelligently generate the machining process of parts by using feature information extracted through feature recognition and a process resource database, thus realizing the intelligent generation of parts machining processes.

[0006] To solve the above problems, the present invention adopts the following technical solution:

[0007] A method for intelligent process design in parts machining, comprising the following steps:

[0008] Step 1: Obtain the MBD 3D model of the part in the 3D software UG, and preprocess the MBD 3D model to obtain the machining feature view, PMI information and process pictures of the MBD 3D model;

[0009] Step 2: Use a processing feature recognizer to identify the processing features in the processing feature view, and display the recognition results in the form of images and text. The processing feature recognizer is obtained by training a deep neural network YOLOv5.

[0010] Step 3: Use the PMI information extraction method to extract process information from the machining features, including the dimensional information, roughness information and tolerance information of the machining features;

[0011] Step 4: Based on PMI information, processing characteristics, and process information, a comprehensive weighted algorithm is used to match processing elements in the process knowledge base. The processing element with the highest matching degree is selected as the processing element corresponding to the processing characteristic. The processing elements of the processing characteristic are then displayed and saved. The formula for the comprehensive weighted algorithm is as follows:

[0012]

[0013] Where, x i The evaluation object includes the type of machining feature, feature size, surface roughness, tolerance grade, and part material. i Indicates the evaluation object x i The corresponding weighting coefficients;

[0014] Step 5: Use the intelligent sorting and optimization algorithm for processing elements to sort all processing elements with processing features and generate the optimal process route. The intelligent sorting and optimization algorithm for processing elements includes the following steps:

[0015] Step 5.1: Determine the number of optimization iterations to begin;

[0016] Step 5.2: Divide all machining elements of the machining features into a reference group and a non-reference group according to the reference-first rule;

[0017] Step 5.3: Based on the roughing-to-finishing rule and the uniqueness rule of the machining element sequence, the benchmark group and non-benchmark group are further subdivided into roughing group, semi-finishing group and finishing group respectively;

[0018] Step 5.4: Based on the rule of surface before hole and the rule of processing layer by layer, divide each group into several level groups;

[0019] Step 5.5: Randomly swap the machining elements within each grade group, and recombine them according to the process rules. Each change of tool, machine tool, and clamping device earns one point. Calculate the iteration score and save the sorting and combination for this time.

[0020] Step 5.6: Decrease the number of optimization iterations by 1, return to step 5.2, and continue until the number of optimization iterations is completed. Take the sorted combination with the lowest score as the optimal process route.

[0021] Step 6: Obtain basic process information through human-computer interaction, including manufacturing unit, part code, and part name, and generate process cards.

[0022] Accordingly, the present invention also proposes an intelligent process design system for parts machining, the system comprising:

[0023] The preprocessing module is used to obtain the MBD 3D model of the part in the 3D software UG, and to preprocess the MBD 3D model to obtain the machining feature view, PMI information and process images of the MBD 3D model.

[0024] The recognition module is used to recognize the processing features of the processing feature view using a processing feature recognizer and to display the recognition results in the form of images and text. The processing feature recognizer is obtained by training the deep neural network YOLOv5.

[0025] The extraction module is used to extract process information from machining features using the PMI information extraction method, including the dimensional information, roughness information, and tolerance information of the machining features;

[0026] The matching module is used to match processing elements in the process knowledge base based on PMI information, processing characteristics, and their technological information using a comprehensive weighted algorithm. The processing element with the highest matching degree is selected as the processing element corresponding to the processing characteristic, and the processing elements of the processing characteristic are displayed and saved. The formula for the comprehensive weighted algorithm is as follows:

[0027]

[0028] Where, x i The evaluation object includes the type of machining feature, feature size, surface roughness, tolerance grade, and part material. i Indicates the evaluation object x i The corresponding weighting coefficients;

[0029] The sorting module is used to sort all processing elements with processing features using a processing element intelligent sorting and optimization algorithm to generate the optimal process route. The processing element intelligent sorting and optimization algorithm includes a determination submodule, a baseline partitioning submodule, a subdivision submodule, a level partitioning submodule, a calculation submodule, and an output submodule.

[0030] The determination submodule is used to determine the number of optimization iterations to begin, and the number is decremented by 1 after each optimization iteration;

[0031] The baseline partitioning submodule is used to divide all machining elements of machining features into baseline groups and non-baseline groups according to the baseline priority rule;

[0032] The subdivision submodule is used to further subdivide the benchmark group and non-benchmark group into roughing group, semi-finishing group and finishing group respectively according to the roughing-to-finishing rule and the unique rule of the machining element sequence;

[0033] The grading submodule is used to gradate each group into several grading groups according to the surface-to-hole rule and the layer-by-layer processing rule of the processing steps.

[0034] The calculated molecular module is used to randomly exchange the machining elements within each level group. After recombination according to the process rules, one point is awarded for each change of the tool, machine tool, and clamping. The iterative score is calculated and the sorting combination is saved.

[0035] The output submodule is used to select the sorted combination with the lowest score as the optimal process route after the optimization iteration loop is completed.

[0036] The generation module is used to obtain basic process information, including manufacturing unit, part code, and part name, through human-computer interaction, and generate process cards.

[0037] The gain effects of this invention are:

[0038] (1) This invention combines part processing feature recognition and intelligent process design to achieve the purpose of communication between design information and manufacturing information, and realizes the effective use of three-dimensional design models for process design.

[0039] (2) This invention effectively integrates the intelligent generation of feature processing method based on part processing feature information and the intelligent sorting of process routes, thereby improving the intelligence of process design;

[0040] (3) The present invention uses intelligent sorting and optimization algorithm to sort the processing elements, and introduces process rules and processing efficiency scoring mechanism, which can quickly obtain efficient and reasonable processing route. Attached Figure Description

[0041] Figure 1 This is a flowchart of the intelligent process design method for part machining according to an embodiment of the present invention;

[0042] Figure 2 This is a diagram of the network structure of the deep neural network YOLOv5;

[0043] Figure 3 A flowchart illustrating the training and feature recognition process of the YOLOv5 deep neural network.

[0044] Figure 4 A flowchart of the PMI information extraction method;

[0045] Figure 5 A flowchart for the intelligent sorting and optimization algorithm for processing elements;

[0046] Figure 6 MBD 3D model diagram of a typical part;

[0047] Figure 7 The image after identifying the machining features of a typical part. Detailed Implementation

[0048] To more clearly illustrate the technical problems, technical solutions, and advantages of the present invention, a detailed description is provided below in conjunction with the accompanying drawings and specific embodiments.

[0049] The intelligent process design method for part machining proposed in this invention first preprocesses the MBD 3D model of the part using UG secondary development technology to obtain the part's feature views, PMI information, and process images. Second, it uses the YOLOv5 algorithm to identify the machining features of the preprocessed MBD 3D model and displays the identification results in the form of images and text. Then, based on the machining features and their process information, a comprehensive weighted algorithm is used to achieve intelligent recommendation of process machining elements. This involves matching instances in a process knowledge base using a comprehensive weighted algorithm, displaying and modifying the optimal matching result, and designing process steps, machine tools, cutting tools, fixtures, and gauges. Finally, using machining elements as the sorting target, an intelligent sorting and optimization algorithm based on process rules and incorporating a machining efficiency scoring mechanism is used to reasonably sort and optimize the machining elements, ensuring that the sorted process routes meet basic process constraints and certain machining quality requirements, ultimately generating the optimal process machining route, i.e., the optimal process route. Finally, a human-computer interaction method is used to obtain some basic information from the process card, including the manufacturing unit, part code, and part name, ultimately generating the process card with one click.

[0050] like Figure 1 As shown, this invention provides an intelligent process design method for part machining based on feature recognition, which mainly includes the following steps:

[0051] Step 1: Preprocess the MBD 3D model of the part, obtain the feature view, PMI information and process images of the MBD 3D model, and generate .png and .avi files respectively;

[0052] Step 2: Automatically identify processing features using a processing feature recognizer based on the deep neural network YOLOv5;

[0053] Step 3: Obtain process information on processing features based on the PMI information extraction method;

[0054] Step 4: Based on the established process knowledge base, using the process information of the part processing features, the processing element of the feature is obtained from the process knowledge base through a comprehensive weighted algorithm, that is, intelligent matching of the processing element corresponding to the processing feature;

[0055] Step 5: Use intelligent sorting and optimization algorithms to sort all the processing elements with processing features in a reasonable and effective manner to generate the optimal process route;

[0056] Step 6: Collect other basic process information through human-computer interaction and generate process cards with one click.

[0057] Step 1 primarily provides the information foundation for other steps. First, the MBD 3D model of the part is obtained in the 3D software UG. Then, the MBD 3D model is preprocessed to obtain the machining feature view, PMI information, and process images of the MBD 3D model.

[0058] MBD (Model-Based Design) 3D models of parts refer to 3D modeling technology used in product design and manufacturing processes. It uses computer software to create virtual models of products, enabling the digitization of design, analysis, simulation, and documentation. These models not only visualize the product but also include product specifications, dimensions, and other design information.

[0059] Feature views refer to the view views of a 3D model captured sequentially based on UG secondary development technology, including the front view, left view, right view, top view, bottom view, and rear view. Figure 6 A basic view.

[0060] The PMI information and process images are obtained through UG secondary development technology, which provides information such as dimensions, roughness, and tolerances on the MBD 3D model of the part. This provides the information basis for subsequent functional operations. The process images are captured using the screenshot function, which provides the information basis for generating the process card in step 6.

[0061] Step 1, the preprocessing of the MBD 3D model using the 3D software UG, specifically includes the following steps:

[0062] First, the machining feature views of the MBD 3D model are captured sequentially using UG secondary development technology. The machining feature views include the front view, left view, right view, top view, bottom view, and rear view.

[0063] Secondly, the PMI information of the MBD 3D model is obtained through UG secondary development technology. The PMI information includes dimensions, roughness and tolerance. The process images of the MBD 3D model are captured using the screenshot function.

[0064] In step 2, a processing feature recognizer is used to identify the processing feature view, and the recognition results are displayed in the form of images and text. The processing feature recognizer is obtained by training the deep neural network YOLOv5.

[0065] YOLOv5 is a single-stage object detection algorithm based on deep neural networks. It is an improvement on the YOLOv4 algorithm, significantly enhancing both detection speed and accuracy. YOLOv5 has four network models of different sizes. YOLOv5s is the smallest network model, while the other three models gradually increase in network depth and width compared to YOLOv5s, with corresponding increases in the number of parameters and computational cost.

[0066] The YOLOv5 network structure diagram in this embodiment is as follows: Figure 2 As shown, the YOLOv5 network structure consists of four parts: input, backbone, neck, and head output.

[0067] At the input end, the model uses Mosaic data augmentation and adaptive anchor box calculation to process the input image, thereby improving the training speed and detection accuracy. At the same time, an adaptive image scaling method is used to scale the input image to adapt to target detection tasks of different sizes and proportions.

[0068] A Focus module is added to the Backbone structure of the feature extraction network to perform two operations on the input image: slicing and channel stitching, in order to reduce the number of network parameters and computational cost.

[0069] The Neck structure utilizes the FPN (Feature Pyramid Network) module and the PAN (Pyramid Attention Network) module to achieve feature fusion of feature maps through upsampling and downsampling, thereby enhancing the feature fusion performance of the network.

[0070] Different loss functions are used at the Head output to calculate the position loss, classification loss, and confidence loss. Feature prediction and classification are achieved through Non-Maximum Suppression (NMS), which improves the network detection accuracy.

[0071] To use the YOLOv5 deep neural network for part machining feature recognition, it is necessary to first train the YOLOv5 deep neural network to obtain a better machining feature recognizer. For example... Figure 3 As shown, the process of training the deep neural network YOLOv5 includes the following steps:

[0072] Step 2.1: First, obtain several machining feature views through UG secondary development technology to generate a machining feature dataset. The machining feature views can be obtained from the MBD 3D models of different parts.

[0073] Step 2.2: Use LabelImg software to annotate each processing feature view in the processing feature dataset. After annotation, it is automatically saved as an XML file, thereby generating a standard XML format dataset, completing the preprocessing of the dataset, and obtaining the preprocessed dataset.

[0074] Step 2.3: Construct the deep neural network YOLOv5, using the preprocessed dataset as the training set for the deep neural network YOLOv5;

[0075] Step 2.4: Continuously train, update parameters and optimize network structure of deep neural network YOLOv5 using the training set. After training, obtain the deep neural network structure with the best recognition performance, and use the deep neural network YOLOv5 with the best recognition performance as the processing feature recognizer.

[0076] After obtaining the machining feature recognizer, the machining feature is identified using the machine feature recognizer. (See also...) Figure 3 The process involves using UG secondary development technology to sequentially capture the machining feature views of the MBD 3D model from step 1, inputting these views into a machining feature recognizer for feature recognition, outputting the recognition results, and displaying the results in both image and text formats. When displaying the recognition results in image and text formats, functions for adding, deleting, modifying, and viewing are included.

[0077] In step 3, the PMI information extraction method is used to extract process information from the machining features. This process information includes the dimensional information, surface roughness information, and tolerance information of the machining features. The PMI information extraction method specifically includes the following steps:

[0078] Step 3.1: Traverse all target information in the MBD 3D model and determine whether the target information has been extracted. If the extraction is complete, end the information extraction; otherwise, proceed to step 3.2. The target information can be any one of the following: size information, roughness information, and tolerance information.

[0079] Step 3.2: Extract all target information from a processing feature;

[0080] Step 3.3: Determine whether the target information of the extracted processing features has been saved into a txt document. If yes, return to step 3.2 and extract all target information of the next processing feature. Otherwise, save the target information of the extracted processing features to a txt document. Stop the loop after all target information of the processing features has been extracted.

[0081] Process information, including dimensions, surface roughness, and tolerances, is extracted from machining features. The following section uses dimensions as an example to illustrate the extraction process; the extraction processes for surface roughness and tolerances are similar. Figure 4 As shown:

[0082] First, iterate through all the dimension information in the MBD 3D model and determine whether all the information has been extracted. If all the dimension information has been extracted, the information extraction ends; otherwise, proceed to the next step.

[0083] Extract all dimensional information from a machining feature;

[0084] Determine whether the extracted dimension information of the processing feature has been saved to a txt document. If the extracted dimension information of the processing feature has not been saved to a txt document, then save it to a txt document. If the extracted dimension information of the processing feature has been saved to a txt document, return to the previous step and re-extract all target information of the next processing feature until the dimension information of all processing features has been extracted, then stop the loop.

[0085] In step 4, based on PMI information, processing features and their process information, a comprehensive weighted algorithm is used to match processing elements in the process knowledge base. The processing element with the highest matching degree is taken as the processing element corresponding to the processing feature, and the processing element of the processing feature is displayed and saved.

[0086] The comprehensive weighted algorithm assigns weights to part processing features, PMI, materials, and other information obtained from upstream steps using the analytic hierarchy process (AHP). Then, a comprehensive weighted formula is used to match processing elements in the process knowledge base, displaying the processing element with the highest matching degree. When displaying processing elements with processing features, a function that allows for manual modification is included.

[0087] Intelligent process generation technology uses a comprehensive weighted algorithm to match feature information and part information in an existing process knowledge base, obtaining the feature machining method with the highest matching degree, thus quickly determining the process step information. The formulas for the comprehensive weighted algorithm are shown in formulas (1) and (2):

[0088]

[0089] Where, x i Indicates the object of evaluation, w i Indicates the evaluation object x iThe corresponding weighting coefficients, where n represents the total number of evaluation objects. The evaluation objects in the above formula include processing feature type, feature size, surface roughness, tolerance grade, and part material. Weights are assigned to the evaluation objects based on expert experience, with weighting coefficients of 50%, 15%, 15%, 10%, and 10% for processing feature type, feature size, surface roughness, tolerance grade, and part material, respectively. The evaluation result Y is compared; a larger value indicates that the corresponding processing method better meets the processing requirements, which is the result recommended by the system.

[0090] In step 5, the intelligent sorting and optimization algorithm for all processing features is used to sort the processing elements and generate the optimal process route.

[0091] The intelligent sorting and optimization algorithm for machining elements mainly adopts process rules and incorporates a scoring mechanism. The process rules include rules such as benchmark first, roughing before finishing, unique machining element sequence, and surface before hole. The machining efficiency is measured by minimizing the number of tool changes, machine tool changes, and clamping changes. A scoring mechanism is added, with one point awarded for each change. The fewer the points, the higher the machining efficiency.

[0092] The rational and efficient sorting of processing elements is achieved through an intelligent sorting and optimization algorithm for processing elements. The flowchart of this algorithm is as follows: Figure 5 As shown, its main steps include:

[0093] Step 5.1: Determine the number N of the initial optimization iterations;

[0094] Step 5.2: Divide all machining elements of machining features into a reference group and a non-reference group according to the reference-first rule (rule 1);

[0095] Step 5.3: Based on the roughing-to-finishing rule (Rule 2) and the unique processing element sequence rule (Rule 3), the benchmark group and non-benchmark group are further subdivided into roughing group, semi-finishing group and finishing group respectively;

[0096] Step 5.4: According to the rule of surface before hole (Rule 4) and the rule of processing layer by layer (Rule 5), classify each group into levels. For example, the level of surface features is higher than that of hole features, and divide them into several level groups.

[0097] Step 5.5: Randomly exchange the machining elements within each grade group, and recombine them according to the process rules. Each change of tool, machine tool, and clamping device earns one point. Calculate the iteration score and save the sorting and combination of this time, i.e., the process route.

[0098] Step 5.6: Decrease the number of optimization iterations i by 1, then return to step 5.2 and repeat the above steps until the number of optimization iterations is completed. Take the sorting combination with the lowest score as the optimal process route.

[0099] In step 6, based on the process images, some basic information on the process card is obtained through human-computer interaction, including the manufacturing unit, part code, part name, etc. Based on the above steps, a PDF process card is generated, including the process route card and the process card, and the process route card and the process card are saved to the local database.

[0100] The invention will be further explained and illustrated below with specific examples.

[0101] When verifying the invention through examples, C++ can be used as the programming language, and a feature recognition-based intelligent process design system can be developed based on the Visual Studio software platform to verify the related technologies and systems through examples.

[0102] The system's functional modules were developed using the Windows 11.0 operating system and Microsoft Visual Studio 2017 programming software. The primary programming language was C / C++, with Qt Toolkit as the application development framework and UG10.0 as an auxiliary tool. MySQL was used as the underlying database. To obtain basic part information, a dynamic link library (.DLL) file generated by the internal development mode was embedded into the UG software using UG secondary development technology. By calling the generated DLL file, the specific function of obtaining basic part information was implemented.

[0103] The following is based on Figure 6 The technical solution of the present invention will be described in detail using the MBD three-dimensional model of a typical part as an example.

[0104] Step 1: Enter the 3D software UG10.0, select the MBD 3D model to be preprocessed, and then use UG secondary development technology to capture the machining feature views of the 3D model in sequence, including the front view, left view, right view, top view, bottom view, and rear view. Figure 6 A basic view. Then, using UG secondary development technology, PMI information such as dimensions, roughness, and tolerances of the MBD 3D model is obtained. The screenshot function is used to capture process images of the 3D model, providing an information basis for subsequent functional operations.

[0105] Step 2: Enter the processing feature recognition page, which displays the six basic views of the 3D model, i.e., .PNG files. Then, the processing features are detected on the feature views, and the detection results are displayed in the form of images and text respectively. Figure 7 Images of typical part processing features are identified, and the part processing features are saved into a database.

[0106] Step 3: Load the .AVI file. The 3D model will automatically perform rotation, flipping and other operations. Then, the machining feature recognizer will automatically identify the features and display the machining feature detection results on the 3D model. Enter the feature recognition result display page, which integrates and processes all the information about the machining features from upstream. The feature information includes name, size, surface roughness and upper and lower tolerances, and has the functions of adding, deleting, modifying and querying.

[0107] Step 4: Fill in the machining information for the part using a human-computer interaction method and save the information into the database. This provides an information source for the design of machining elements and content assistance for the generation of process cards. Next, display the feature information in the part machining feature database. Then, right-click on each feature information to bring up the machining element design page. Finally, after completing the machining element design, close the page. The feature information will automatically turn blue to remind the operator that the machining element design for this feature has been completed, avoiding duplicate design.

[0108] Step 5: Upon entering the machining element design page, feature information and part size information will be automatically displayed, and feature tolerances will be calculated, providing an information basis for the design of feature machining elements. The machining steps with the highest matching degree will be displayed. Then, for each machining step, the tool with the highest similarity will be matched, displaying the tool model, tool name, and detailed tool information. The matching operation for machine tools, fixtures, and gauges is the same as for tools. Finally, the feature machining element design is completed, and all machining element information is saved into the database.

[0109] Step 6: Fill in some basic information about part generation. The system will automatically calculate the production plan and display the specific information of the part, as well as recommend the selection of process principles.

[0110] Step 7: Sort the processing elements according to the process rules to form a reasonable process route. Input the number of iterations, and use a scoring mechanism to re-sort the steps to improve the processing efficiency of the process route. Manual modification is also included to further refine the process route. Finally, the steps are integrated into processes, and the processes are automatically labeled. The process of distributing processes is handled similarly.

[0111] Step 8: The system will automatically fill in the process card content and output the process route diagram, arrange all process cards according to the process sequence, and finally save the process cards locally. The file name can be modified.

[0112] In another embodiment, the present invention also provides an intelligent process design system for parts processing, the system comprising a preprocessing module, an identification module, an extraction module, a matching module, a sorting module, and a generation module.

[0113] The preprocessing module is used to obtain the MBD 3D model of the part in the 3D software UG, and to preprocess the MBD 3D model to obtain the machining feature view, PMI information and process images of the MBD 3D model.

[0114] The recognition module is used to identify the processing features of the processing feature view using a processing feature recognizer and to display the recognition results in the form of images and text. The processing feature recognizer is obtained by training the deep neural network YOLOv5.

[0115] The extraction module is used to extract process information from machining features using the PMI information extraction method, including the dimensional information, roughness information and tolerance information of the machining features.

[0116] The matching module is used to match processing elements in the process knowledge base based on PMI information, processing characteristics, and process information using a comprehensive weighted algorithm. The processing element with the highest matching degree is selected as the processing element corresponding to the processing characteristic, and the processing elements of the processing characteristic are displayed and saved. The formula of the comprehensive weighted algorithm is as follows:

[0117]

[0118] Where, x i Indicates the object of evaluation, w i Indicates the evaluation object x i The corresponding weighting coefficients, where n represents the total number of evaluation objects. The evaluation objects in the above formula include processing feature type, feature size, surface roughness, tolerance grade, and part material. Weights are assigned to the evaluation objects based on expert experience, with weighting coefficients of 50%, 15%, 15%, 10%, and 10% for processing feature type, feature size, surface roughness, tolerance grade, and part material, respectively. The evaluation result Y is compared; a larger value indicates that the corresponding processing method better meets the processing requirements, which is the result recommended by the system.

[0119] The sorting module is used to sort all machining elements with machining features using a machining element intelligent sorting and optimization algorithm to generate the optimal process route. The machining element intelligent sorting and optimization algorithm includes a determination submodule, a baseline partitioning submodule, a subdivision submodule, a level partitioning submodule, a calculation submodule, and an output submodule.

[0120] The determination submodule is used to determine the number of optimization iterations to begin, and the number is decremented by 1 after each optimization iteration;

[0121] The baseline partitioning submodule is used to divide all machining elements of machining features into baseline groups and non-baseline groups according to the baseline priority rule;

[0122] The subdivision submodule is used to further subdivide the benchmark group and non-benchmark group into roughing group, semi-finishing group and finishing group respectively according to the roughing-to-finishing rule and the unique rule of the machining element sequence;

[0123] The grading submodule is used to gradate each group into several grading groups according to the surface-to-hole rule and the layer-by-layer processing rule of the processing steps.

[0124] The calculated molecular module is used to randomly exchange the machining elements within each level group. After recombination according to the process rules, one point is awarded for each change of the tool, machine tool, and clamping. The iterative score is calculated and the sorting combination is saved.

[0125] The output submodule is used to select the sorting combination with the lowest score as the optimal process route after the optimization iteration loop is completed.

[0126] The generation module is used to obtain basic process information through human-computer interaction, including manufacturing unit, part code, and part name, generate process cards, and save the process cards to the local database.

[0127] The specific implementation methods of each module in the intelligent process design system for part processing of the present invention can refer to the implementation methods described in the above embodiments of the intelligent process design method for part processing, and will not be repeated here.

[0128] This invention combines part processing feature recognition and intelligent process design, achieving the goal of communication between design information and manufacturing information, and realizing the effective use of three-dimensional design models for process design.

[0129] This invention effectively integrates a method for intelligently generating feature processing based on part processing feature information and an intelligent sorting of process routes, thereby improving the intelligence of process design.

[0130] This invention employs an intelligent sorting and optimization algorithm for processing elements to sort them, and introduces process rules and a processing efficiency scoring mechanism, which can quickly obtain efficient and reasonable processing routes.

[0131] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0132] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are relatively specific and detailed, they should not be construed as limiting the scope of the invention patent. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these all fall within the protection scope of the present invention. Therefore, the protection scope of this invention patent should be determined by the appended claims.

Claims

1. A method for intelligent process design in parts machining, characterized in that, Includes the following steps: Step 1: Obtain the MBD 3D model of the part in the 3D software UG, and preprocess the MBD 3D model to obtain the machining feature view, PMI information and process pictures of the MBD 3D model; Step 2: Use a processing feature recognizer to identify the processing features in the processing feature view, and display the recognition results in the form of images and text. The processing feature recognizer is obtained by training a deep neural network YOLOv5. Step 3: Use the PMI information extraction method to extract process information from the machining features, including the dimensional information, roughness information and tolerance information of the machining features; Step 4: Based on PMI information, processing characteristics, and process information, a comprehensive weighted algorithm is used to match processing elements in the process knowledge base. The processing element with the highest matching degree is selected as the processing element corresponding to the processing characteristic. The processing elements of the processing characteristic are then displayed and saved. The formula for the comprehensive weighted algorithm is as follows: Where, x i The evaluation object includes the type of machining feature, feature size, surface roughness, tolerance grade, and part material. i Indicates the evaluation object x i The corresponding weighting coefficients; Step 5: Use the intelligent sorting and optimization algorithm for processing elements to sort all processing elements with processing features and generate the optimal process route. The intelligent sorting and optimization algorithm for processing elements includes the following steps: Step 5.1: Determine the number of optimization iterations to begin; Step 5.2: Divide all machining elements of the machining features into a reference group and a non-reference group according to the reference-first rule; Step 5.3: Based on the roughing-to-finishing rule and the unique processing element sequence rule, the benchmark group and non-benchmark group are further subdivided into roughing group, semi-finishing group and finishing group respectively; Step 5.4: Based on the rule of surface before hole and the rule of processing layer by layer, divide each group into several level groups; Step 5.5: Randomly swap the machining elements within each grade group, and recombine them according to the process rules. Each change of tool, machine tool, and clamping device earns one point. Calculate the iteration score and save the sorting and combination for this time. Step 5.6: Decrease the number of optimization iterations by 1, return to step 5.2, and continue until the number of optimization iterations is completed. Take the sorted combination with the lowest score as the optimal process route. Step 6: Obtain basic process information through human-computer interaction, including manufacturing unit, part code, and part name, and generate process cards.

2. The intelligent process design method for part machining according to claim 1, characterized in that, The preprocessing process for MBD 3D models includes: The machining feature views of the MBD 3D model are captured sequentially using UG secondary development technology, including the front view, left view, right view, top view, bottom view, and rear view. The PMI information of the MBD 3D model, including dimensions, roughness, and tolerances, was obtained through UG secondary development technology, and the process images of the MBD 3D model were captured using the screenshot function.

3. The intelligent process design method for part machining according to claim 1 or 2, characterized in that, The YOLOv5 deep neural network includes an input terminal, a backbone structure, a neck structure, and a head output terminal. The input image is processed using Mosaic data augmentation and adaptive anchor box calculation methods, and is also scaled using an adaptive image scaling method. Add a Focus module to the Backbone structure to perform two operations on the input image: slicing and channel stitching. In the Neck structure, feature fusion of feature maps is achieved by using the FPN and PAN modules for upsampling and downsampling. Different loss functions are used at the Head output to calculate the location loss, classification loss, and confidence loss, and feature prediction and classification are achieved through NMS (non-maximum suppression).

4. The intelligent process design method for part machining according to claim 3, characterized in that, The process of training the deep neural network YOLOv5 includes the following steps: Step 2.1: Obtain several machining feature views using UG secondary development technology to generate a machining feature dataset; Step 2.2: Use LabelImg software to annotate each processing feature view in the processing feature dataset. After annotation, it is automatically saved as an XML file, thereby generating a standard XML format dataset and obtaining the preprocessed dataset. Step 2.3: Construct the deep neural network YOLOv5, using the preprocessed dataset as the training set for the deep neural network YOLOv5; Step 2.4: Continuously train the deep neural network YOLOv5 using the training set, update the parameters and optimize the network structure, and use the deep neural network YOLOv5 with the best recognition performance after training as the processing feature recognizer.

5. The intelligent process design method for part machining according to claim 1 or 2, characterized in that, The PMI information extraction method includes the following steps: Step 3.1: Traverse all target information in the MBD 3D model and determine whether the target information has been extracted. If the extraction is complete, end the information extraction; otherwise, proceed to step 3.

2. The target information can be any one of the following: size information, roughness information, and tolerance information. Step 3.2: Extract all target information from a processing feature; Step 3.3: Determine whether the target information of the extracted processing features has been saved into a txt document. If yes, return to step 3.2 and extract all target information of the next processing feature. Otherwise, save the target information of the extracted processing features to a txt document. Stop the loop after all target information of the processing features has been extracted.

6. The intelligent process design method for part machining according to claim 1 or 2, characterized in that, The weighting coefficients for machining feature type, feature size, surface roughness, tolerance grade, and part material are 50%, 15%, 15%, 10%, and 10%, respectively.

7. The intelligent process design method for part machining according to claim 1 or 2, characterized in that, The processing feature views of the MBD 3D model include the front view, left view, right view, top view, bottom view, and rear view.

8. The intelligent process design method for part machining according to claim 1 or 2, characterized in that, When displaying the recognition results in the form of images and text, add functions for adding, deleting, modifying, and viewing.

9. The intelligent process design method for part machining according to claim 1 or 2, characterized in that, When displaying the processing elements of the processing features, a function that allows manual modification is added.

10. An intelligent process design system for parts machining, characterized in that, include: The preprocessing module is used to obtain the MBD 3D model of the part in the 3D software UG, and to preprocess the MBD 3D model to obtain the machining feature view, PMI information and process images of the MBD 3D model. The recognition module is used to identify the processing features of the processing feature view using a processing feature recognizer and to display the recognition results in the form of images and text. The processing feature recognizer is obtained by training the deep neural network YOLOv5. The extraction module is used to extract process information from machining features using the PMI information extraction method, including the dimensional information, roughness information, and tolerance information of the machining features; The matching module is used to match processing elements in the process knowledge base based on PMI information, processing characteristics, and their technological information using a comprehensive weighted algorithm. The processing element with the highest matching degree is selected as the processing element corresponding to the processing characteristic, and the processing elements of the processing characteristic are displayed and saved. The formula for the comprehensive weighted algorithm is as follows: Where, x i The evaluation object includes the type of machining feature, feature size, surface roughness, tolerance grade, and part material. i Indicates the evaluation object x i The corresponding weighting coefficients; The sorting module is used to sort all processing elements with processing features using a processing element intelligent sorting and optimization algorithm to generate the optimal process route. The processing element intelligent sorting and optimization algorithm includes a determination submodule, a baseline partitioning submodule, a subdivision submodule, a level partitioning submodule, a calculation submodule, and an output submodule. The determination submodule is used to determine the number of optimization iterations to begin, and the number is decremented by 1 after each optimization iteration; The baseline partitioning submodule is used to divide all machining elements of machining features into baseline groups and non-baseline groups according to the baseline priority rule; The subdivision submodule is used to further subdivide the benchmark group and non-benchmark group into roughing group, semi-finishing group and finishing group respectively according to the roughing-to-finishing rule and the unique rule of the machining element sequence; The grading submodule is used to gradate each group into several grading groups according to the surface-to-hole rule and the layer-by-layer processing rule of the processing steps. The calculated molecular module is used to randomly exchange the machining elements within each level group. After recombination according to the process rules, one point is awarded for each change of the tool, machine tool, and clamping. The iterative score is calculated and the sorting combination is saved. The output submodule is used to select the sorted combination with the lowest score as the optimal process route after the optimization iteration loop is completed. The generation module is used to obtain basic process information, including manufacturing unit, part code, and part name, through human-computer interaction, and generate process cards.

Images