CN116976034A - CAD software-based part library system - Google Patents
CAD software-based part library system Download PDFInfo
- Publication number
- CN116976034A CN116976034A CN202311002873.1A CN202311002873A CN116976034A CN 116976034 A CN116976034 A CN 116976034A CN 202311002873 A CN202311002873 A CN 202311002873A CN 116976034 A CN116976034 A CN 116976034A
- Authority
- CN
- China
- Prior art keywords
- module
- parts
- unit
- cad
- input
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000013461 design Methods 0.000 claims abstract description 52
- 230000006870 function Effects 0.000 claims description 51
- 238000007726 management method Methods 0.000 claims description 31
- 238000000034 method Methods 0.000 claims description 30
- 238000012549 training Methods 0.000 claims description 30
- 238000004458 analytical method Methods 0.000 claims description 28
- 238000004422 calculation algorithm Methods 0.000 claims description 26
- 238000011176 pooling Methods 0.000 claims description 22
- 238000001514 detection method Methods 0.000 claims description 19
- 238000013527 convolutional neural network Methods 0.000 claims description 16
- 230000011218 segmentation Effects 0.000 claims description 14
- 238000000605 extraction Methods 0.000 claims description 13
- 230000010354 integration Effects 0.000 claims description 13
- 230000008439 repair process Effects 0.000 claims description 13
- 238000012545 processing Methods 0.000 claims description 12
- 239000000284 extract Substances 0.000 claims description 10
- 238000013499 data model Methods 0.000 claims description 9
- 238000010586 diagram Methods 0.000 claims description 9
- 230000001360 synchronised effect Effects 0.000 claims description 8
- 230000005540 biological transmission Effects 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000005070 sampling Methods 0.000 claims description 6
- 230000008520 organization Effects 0.000 claims description 5
- 230000004913 activation Effects 0.000 claims description 4
- 238000013528 artificial neural network Methods 0.000 claims description 4
- 238000004364 calculation method Methods 0.000 claims description 4
- 230000008859 change Effects 0.000 claims description 4
- 238000006243 chemical reaction Methods 0.000 claims description 4
- 238000013480 data collection Methods 0.000 claims description 4
- 238000007405 data analysis Methods 0.000 claims description 3
- 230000004927 fusion Effects 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 238000012986 modification Methods 0.000 claims description 3
- 230000004048 modification Effects 0.000 claims description 3
- 230000000306 recurrent effect Effects 0.000 claims description 3
- 238000002372 labelling Methods 0.000 claims 1
- 238000011960 computer-aided design Methods 0.000 abstract description 3
- 230000008569 process Effects 0.000 description 8
- 238000004519 manufacturing process Methods 0.000 description 7
- 238000012216 screening Methods 0.000 description 4
- 230000007547 defect Effects 0.000 description 3
- 238000005516 engineering process Methods 0.000 description 3
- 239000000463 material Substances 0.000 description 3
- 238000003058 natural language processing Methods 0.000 description 3
- 238000012795 verification Methods 0.000 description 3
- 230000008485 antagonism Effects 0.000 description 2
- 238000012938 design process Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 230000018109 developmental process Effects 0.000 description 2
- 230000008676 import Effects 0.000 description 2
- 238000007689 inspection Methods 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 230000007246 mechanism Effects 0.000 description 2
- 238000010845 search algorithm Methods 0.000 description 2
- 238000012360 testing method Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008901 benefit Effects 0.000 description 1
- 125000004122 cyclic group Chemical group 0.000 description 1
- 238000013500 data storage Methods 0.000 description 1
- 230000002950 deficient Effects 0.000 description 1
- 238000009826 distribution Methods 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 239000003607 modifier Substances 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012800 visualization Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/17—Mechanical parametric or variational design
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/0464—Convolutional networks [CNN, ConvNet]
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/048—Activation functions
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/764—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using classification, e.g. of video objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/82—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using neural networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- Software Systems (AREA)
- General Health & Medical Sciences (AREA)
- Computing Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Geometry (AREA)
- Mathematical Physics (AREA)
- Molecular Biology (AREA)
- Data Mining & Analysis (AREA)
- Computational Linguistics (AREA)
- Biophysics (AREA)
- Biomedical Technology (AREA)
- Databases & Information Systems (AREA)
- Multimedia (AREA)
- Medical Informatics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Computer Hardware Design (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
Abstract
The invention discloses a CAD software-based part library system, which relates to the field of computer-aided design and comprises a CAD integrated module, a part filing module, a part standardization module, an intelligent retrieval module, a part version management module, a part community module and an AI collaborative design module, wherein the output end of the part filing module is connected with the input end of the part standardization module, and the output end of the part standardization module is connected with the input end of the intelligent retrieval module; the intelligent retrieval module output end is connected with the CAD integrated module input end, the part version management control module output end is connected with the part archiving module input end, and the AI collaborative design module output end is connected with the CAD integrated module input end.
Description
Technical Field
The present invention relates to the field of computer aided design, and more particularly to a CAD software-based parts library system.
Background
With the rapid development of digital and smart manufacturing, the traditional manufacturing industry is undergoing a revolutionary revolution. In the past few years, the use of computer aided design CAD technology in the field of industrial manufacturing has been widely spread and applied. Conventional parts library systems are typically managed based on paper documents and manual recordings, which do not meet the requirements of rapid changes and efficient production. However, with the advent and development of CAD software, manufacturers can design, develop, and manage components digitally. CAD software provides powerful design and simulation tools that allow engineers to design and simulate parts on a computer. This greatly improves the efficiency and accuracy of the design. In addition, CAD software also supports the import and export of various file formats, which is convenient for data intercommunication with other systems. The CAD software-based part library system utilizes the advantages of a computer to realize the functions of inventory management, document management, version control and the like of parts on the basis of digital design. Through integrating CAD software and a part library system, enterprises can realize comprehensive digital management, and production efficiency and product quality are improved. However, conventional CAD software-based parts library systems have some drawbacks:
Conventional parts library systems typically employ manual entry and management, requiring significant time and effort by personnel to maintain part information. This makes the system inefficient and prone to errors and re-operation.
The addition of new parts to conventional parts library systems typically requires manual intervention, including filling in detailed information, uploading related files, and the like. This makes the part addition process cumbersome and time consuming, and prone to errors or omissions.
Conventional parts library systems often lack good collaboration and sharing mechanisms. Part information cannot be shared and accessed in real time among different departments, so that cooperation is difficult and information is isolated. Furthermore, the interoperability of conventional systems is more limited for cross-department or cross-organization collaboration, and it is difficult to achieve efficient collaboration.
In general, conventional parts library systems have some significant drawbacks in terms of efficiency, parts addition, and interoperability, requiring the introduction of new techniques and methods to address these issues.
Accordingly, the present invention discloses a CAD software-based part library system.
Disclosure of Invention
Aiming at the defects of the prior art, the invention discloses a CAD software-based part library system, which tightly combines the design with the part library system through seamless integration with CAD software; the complex process of manual input and repeated work is eliminated, and the efficiency and accuracy of part addition are improved. The designed parts are archived and stored according to the technical properties and classification by a part archiving module; and the problems of information isolation and repeated design in the traditional system are avoided. By standardizing the module and managing the design and specification of the parts, errors and risks in design and production are reduced. The intelligent retrieval module utilizes advanced search algorithm and semantic analysis technology to realize efficient and accurate part retrieval; the time for searching and screening the parts is shortened, and the efficiency is improved. Recording and managing different version information of the part through a part version management module; avoiding outdated and conflicting problems. Providing an exchange and sharing platform for engineers through a part community module; the problem of poor collaboration is solved. The AI collaborative design module is used for providing intelligent assistance and support for the design process; improving the efficiency and quality of the design and facilitating collaboration and innovation among design teams.
The invention adopts the following technical scheme:
a CAD software-based parts library system, the system comprising:
the CAD integration module integrates the part library system with CAD software through a CAD fusion plug-in unit so as to directly access and use components in the part library in the CAD software;
the part archiving module automatically identifies parts and rapidly classifies the parts in a grading way through the data collecting unit, the model training unit, the image identification classifying unit, the part information extracting unit and the automatic archiving unit;
the part standardization module automatically detects and verifies the parameter consistency of the parts in engineering design through a model detection and geometric analysis algorithm;
the intelligent search module is used for realizing an intelligent search function through an intelligent language model based on the part archiving module data; the intelligent language model automatically matches and recommends corresponding parts by analyzing keywords and descriptions input by a user, and provides associated classifications and labels;
the part version management module tracks and manages the change history and version information of the part through the metadata management unit and the version control unit and merges the part data;
The part community module establishes an online part sharing platform through a cloud server erection method so as to promote knowledge exchange and innovation cooperation among users;
the AI collaborative design module analyzes the design intent of the user through the intelligent language model and provides design choices and optimization suggestions through generating an countermeasure network model;
the output end of the part filing module is connected with the input end of the part standardization module, and the output end of the part standardization module is connected with the input end of the intelligent retrieval module; the output end of the intelligent search module is connected with the input end of the CAD integrated module, the output end of the part version management control module is connected with the input end of the part archiving module, and the output end of the AI collaborative design module is connected with the input end of the CAD integrated module.
As a further technical scheme of the invention, the CAD integrated module comprises a data transmission unit, a user interface expansion unit, a CAD file processing unit and a synchronous updating unit; the data transmission unit acquires part metadata information from the part library system through an application program interface API; and transmitting the data to the CAD software via the application program interface API to access and use the part directly in the CAD; the user interface expansion unit adds a part library search column, a part preview image and a component selector in CAD software by introducing an interface element tool kit so as to facilitate browsing and selecting components in the part library; the CAD file processing unit realizes the importing and exporting functions of CAD file formats through a file format conversion tool so as to facilitate archiving management; when the components in the part library are changed, the CAD integration module updates the component list in the CAD software in real time through the synchronous updating unit, so that the user can always access the latest part information.
As a further technical scheme of the invention, the data collection unit collects a large number of part images and corresponding label information through an Internet retrieval tool, and a training data set is established; the model training unit trains the training data set through a convolutional neural network algorithm so as to learn the characteristics and the types of the parts; the image recognition and classification unit scans images through a scanner and automatically recognizes and classifies parts by using a trained model, part information is automatically extracted from the Internet through the part information extraction unit after recognition and classification are completed, and the part information extraction unit analyzes and extracts the acquired information through a data analysis tool; after the information extraction and analysis are completed, automatically archiving the parts subjected to identification and classification into a data model through an automatic archiving unit so as to facilitate searching and management; the automatic archiving unit records the categories and labels of the parts through a metadata recording tool to support the organization and retrieval of the files.
As a further technical scheme of the invention, the convolutional neural network algorithm extracts features from an input image through a convolutional function, a small area of the input image is given as a convolutional kernel and a weight matrix, the convolutional operation multiplies the convolutional kernel and a local area of the image element by element, and the whole image is traversed through a window of the sliding convolutional kernel; the expression of the convolution function F is as follows,
In the formula (1), M is one pixel point in the output feature map of the convolution operation, β is one pixel point of the input image, D is the weight of the convolution kernel, and g is the bias term; after the convolution is completed and the stacking is completed, the number of image pixel blocks is excessive, the size of the feature map is reduced through maximum pooling, the calculation complexity is reduced, and the main features are extracted; the maximum pooling function divides the input area into non-overlapping subareas, then selects the maximum value in each subarea as output, the formula of the maximum pooling function is that,
in the formula (2), b is a pixel point in an output characteristic diagram of the pooling operation, t is a sub-region in an input characteristic diagram, and θ is a pooling step size; multiple convolution layers, activation functions and pooling layers are repeatedly stacked in the maximum pooling to learn the details and abstract features of the input image and conduct accurate classification prediction.
As a further technical scheme of the invention, the part standardization module comprises a parameter detection unit, a standard matching unit and an error repair unit; the parameter detection unit reads and analyzes CAD files of the parts through CAD software, and compares whether the parts meet the requirements through a logic judgment algorithm; the standard matching unit compares the parameters and the characteristics of the parts with the matching degree of the standards through the matching rules, and when the parts are detected to be out of compliance with the standards, the error repairing unit automatically repairs the errors of the parts.
As a further technical scheme of the invention, the metadata management unit stores and organizes information and attribute data of the parts through a part library database; the part library database stores unique identifiers, version numbers, attributes and CAD file path information of each part and comprises creation date, modification date, author, description and relation with other parts; the version control unit tracks and manages different versions of the parts through version management software SVN, and performs comparison and merging operations between the versions.
As a further technical solution of the present invention, the generating an antagonism network model generates antagonism samples through a generator and a discriminator to provide design choices and optimization suggestions, and the generator generates falsified samples by using random input as input data through the generating model; the discriminator distinguishes the fake sample and the real sample generated by the generator through the discrimination model, the generation model and the discrimination model are trained alternately through an countermeasure training method, and network parameters are updated through a counter propagation algorithm, so that the generator is better in generating the real sample, and the discriminator is better in distinguishing the real sample and the fake sample; the method for generating the countermeasure network model countermeasure training comprises the following steps of;
Step 1, randomly sampling from real data to serve as an input sample of a discriminator, and carrying out forward propagation of the discriminator to obtain a discrimination result of the discriminator on the real sample;
step 2, sampling from random input as an input sample of a generator, and performing forward propagation of the generator to generate a forged sample;
step 3, taking the sample generated by the generator and the real sample as the input of the discriminator, and carrying out forward propagation of the discriminator to obtain the discrimination result of the discriminator on the forged sample and the real sample;
step 4, calculating a loss function of the discriminator according to a discrimination result of the discriminator, and carrying out back propagation to update parameters of the discriminator;
step 5, calculating a loss function of the generator according to a discrimination result obtained by a discriminator of the sample generated by the generator, and carrying out back propagation to update parameters of the generator;
and step 6, repeating the steps until the preset training iteration times or the convergence of the loss function are reached.
As a further technical scheme of the invention, the model detection is based on a pre-constructed three-dimensional model and corresponding two-dimensional projection information, the shape similarity among parts is calculated through a part shape similarity function, the parameter consistency is verified, the part shape similarity function expression is as follows,
S＝∑(α×A 1 +δ×A 2 +γ×A 3 ) 2 (3)
In the formula (3), S represents the shape similarity of the parts, A 1 、A 2 、A 3 Characteristic parameters of the part are represented, and alpha, delta and gamma are weight coefficients; the consistency of parameters is verified by comparing the shape similarity of different parts in the formula (3); if the shape similarity of the different parts is equal, the parameters are consistent;
the geometric analysis algorithm verifies whether the size parameters of different parts are consistent through the consistency of the size parameters; the dimensional parameter consistency is mathematically expressed as follows:
in the formula (4), P represents a dimensional parameter of the part, V represents an average value of the dimensional parameter, and ε is an allowable error threshold.
As a further technical scheme of the invention, the intelligent language model analyzes user input and design intent through a word splitting module, a part-of-speech tagging module, an entity recognition module and a syntactic analysis module; the word segmentation module segments the text input by the user through the existing rule word segmentation method, and the word segmentation module is used for understanding sentence meaning; the part-of-speech tagging module predicts part-of-speech tags corresponding to words through a conditional random field model; the entity recognition module learns and retrieves the recognition word expression entity through a recurrent neural network and marks the recognition word expression entity in a classified manner; the syntactic analysis module analyzes the dependency relationship among the words through a dependency syntactic analyzer to infer the grammar structure of sentences and give analysis values, and the output end of the word segmentation module is connected with the input end of the part-of-speech tagging module; the output end of the part-of-speech tagging module is connected with the input end of the entity identification module; the output end of the entity identification module is connected with the input end of the syntactic analysis module.
Has the positive beneficial effects that:
the invention discloses a CAD software-based part library system, which is used for tightly combining a design with the part library system through seamless integration with CAD software; the complex process of manual input and repeated work is eliminated, and the efficiency and accuracy of part addition are improved. The designed parts are archived and stored according to the technical properties and classification by a part archiving module; and the problems of information isolation and repeated design in the traditional system are avoided. By standardizing the module and managing the design and specification of the parts, errors and risks in design and production are reduced. The intelligent retrieval module utilizes advanced search algorithm and semantic analysis technology to realize efficient and accurate part retrieval; the time for searching and screening the parts is shortened, and the efficiency is improved. Recording and managing different version information of the part through a part version management module; avoiding outdated and conflicting problems. Providing an exchange and sharing platform for engineers through a part community module; the problem of poor collaboration is solved. The AI collaborative design module is used for providing intelligent assistance and support for the design process; improving the efficiency and quality of the design and facilitating collaboration and innovation among design teams.
Drawings
FIG. 1 is a schematic flow diagram of an overall module of a CAD software-based part library system of the present invention;
FIG. 2 is a schematic diagram of a CAD software-based part library system generation countermeasure network of the present invention;
FIG. 3 is a training structure diagram of a convolutional neural network algorithm in a CAD software-based part library system according to the present invention;
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
A CAD software-based parts library system, the system comprising:
the CAD integration module integrates the part library system with CAD software through a CAD fusion plug-in unit so as to directly access and use components in the part library in the CAD software;
the part archiving module automatically identifies parts and rapidly classifies the parts in a grading way through the data collecting unit, the model training unit, the image identification classifying unit, the part information extracting unit and the automatic archiving unit;
The part standardization module automatically detects and verifies the parameter consistency of the parts in engineering design through a model detection and geometric analysis algorithm;
the intelligent search module is used for realizing an intelligent search function through an intelligent language model based on the part archiving module data; the intelligent language model automatically matches and recommends corresponding parts by analyzing keywords and descriptions input by a user, and provides associated classifications and labels;
the part version management module tracks and manages the change history and version information of the part through the metadata management unit and the version control unit and merges the part data;
the part community module establishes an online part sharing platform through a cloud server erection method so as to promote knowledge exchange and innovation cooperation among users;
the AI collaborative design module analyzes the design intent of the user through the intelligent language model and provides design choices and optimization suggestions through generating an countermeasure network model;
the output end of the part filing module is connected with the input end of the part standardization module, and the output end of the part standardization module is connected with the input end of the intelligent retrieval module; the output end of the intelligent search module is connected with the input end of the CAD integrated module, the output end of the part version management control module is connected with the input end of the part archiving module, and the output end of the AI collaborative design module is connected with the input end of the CAD integrated module.
In the above embodiment, the CAD integration module includes a data transmission unit, a user interface expansion unit, a CAD file processing unit, and a synchronization update unit; the data transmission unit acquires part metadata information from the part library system through an application program interface API; and transmitting the data to the CAD software via the application program interface API to access and use the part directly in the CAD; the user interface expansion unit adds a part library search column, a part preview image and a component selector in CAD software by introducing an interface element tool kit so as to facilitate browsing and selecting components in the part library; the CAD file processing unit realizes the importing and exporting functions of CAD file formats through a file format conversion tool so as to facilitate archiving management; when the components in the part library are changed, the CAD integration module updates the component list in the CAD software in real time through the synchronous updating unit, so that the user can always access the latest part information.
In a specific embodiment, the CAD integration module achieves the function of acquiring the metadata information of the parts from the parts library system by establishing a data connection with the parts library system or using a data exchange format (such as XML or JSON). And allows the CAD software to interact with the part library system through the application program API to obtain the required part information. Meanwhile, the CAD integration module uses a graphical interface toolkit (such as Qt or WPF) to create a custom interface element, embeds the custom interface element into CAD software, and adds the functions of a part library search bar, a part preview image, a component selector and the like by expanding the interface of the CAD software. In addition, the CAD integration module realizes the importing and exporting functions of the CAD file format through a CAD file processing library or an API. The CAD file is converted into a format supported by the parts library system by reading and analyzing the format of the CAD file. In this way, CAD files can be archived in the parts library system and re-imported into CAD software for editing and use as needed. When the components in the part library are changed, the component list in the CAD software is updated in real time through the synchronous updating unit. The synchronous updating unit detects the change in the parts library system by periodically polling or using a mechanism such as a message queue, and transmits new parts information to the CAD software for updating. This ensures that the user has constant access to the latest part information and avoids the problem of inconsistent data.
In the above embodiment, the data collection unit collects a large number of part images and corresponding tag information through an internet search tool, and establishes a training data set; the model training unit trains the training data set through a convolutional neural network algorithm so as to learn the characteristics and the types of the parts; the image recognition and classification unit scans images through a scanner and automatically recognizes and classifies parts by using a trained model, part information is automatically extracted from the Internet through the part information extraction unit after recognition and classification are completed, and the part information extraction unit analyzes and extracts the acquired information through a data analysis tool; after the information extraction and analysis are completed, automatically archiving the parts subjected to identification and classification into a data model through an automatic archiving unit so as to facilitate searching and management; the automatic archiving unit records the categories and labels of the parts through a metadata recording tool to support the organization and retrieval of the files.
In a specific embodiment, the data collection unit collects a large number of part images and corresponding label information from various online sources using an internet search tool. These images and label information are organized to create a rich and diverse training data set. The model training unit then processes and trains this training dataset. The convolutional neural network algorithm is applied to the training process, and features of the parts can be automatically extracted by learning the part images, and categories of different parts can be learned. In practical application, the image recognition and classification unit scans the part images by placing the part images in a scanner, and then automatically recognizes and classifies the part images by using a model trained before. Once the identification classification is completed, the part information extraction unit starts to automatically extract the relevant information of the part from the internet. This process may require parsing, extraction, and further organization of information obtained from the web page for subsequent processing by means of a data parsing tool. After the analysis and extraction are completed, the automatic filing unit is responsible for automatically filing the identified and classified parts into the data model. The data model may be a database or other suitable data structure to facilitate the searching and management of the parts. Meanwhile, the automatic filing unit also uses a metadata recording tool to record the category and label information of each part, so that the organization and the retrieval of files can be supported, and a user can conveniently and quickly find the required part.
In summary, the system collects data through an internet retrieval tool, performs training by applying a convolutional neural network algorithm, recognizes and classifies the data through a scanner, extracts information from the internet, automatically files the data into a data model, records metadata and the like, and realizes a complete process from part image to automatic file.
In the above embodiment, the convolutional neural network algorithm extracts features from the input image through a convolutional function, and gives a small area of the input image as a convolutional kernel and a weight matrix, and the convolutional operation multiplies the convolutional kernel by the local area of the image element by element, and traverses the whole image by sliding a window of the convolutional kernel; the expression of the convolution function F is as follows,
in the formula (1), M is one pixel point in the output feature map of the convolution operation, β is one pixel point of the input image, D is the weight of the convolution kernel, and g is the bias term; after the convolution is completed and the stacking is completed, the number of image pixel blocks is excessive, the size of the feature map is reduced through maximum pooling, the calculation complexity is reduced, and the main features are extracted; the maximum pooling function divides the input area into non-overlapping subareas, then selects the maximum value in each subarea as output, the formula of the maximum pooling function is that,
In the formula (2), b is a pixel point in an output characteristic diagram of the pooling operation, t is a sub-region in an input characteristic diagram, and θ is a pooling step size; multiple convolution layers, activation functions and pooling layers are repeatedly stacked in the maximum pooling to learn the details and abstract features of the input image and conduct accurate classification prediction.
In a specific embodiment, as shown in fig. 3, the convolutional neural network algorithm creates a training dataset comprising a plurality of part images and corresponding labels by acquiring design images of the parts from CAD software and performing fine-tuning of parameter settings. And performing feature extraction processing on the training data set, and performing binary conversion on the image features to ensure consistency and expandability of input data. And according to the task requirements and the characteristics of the data set, a proper convolutional neural network architecture is designed to perform primary screening search and accurate screening search. Typical architectures may include convolutional layers, activation functions, pooling layers, fully-connected layers, and the like. Meanwhile, the convolutional neural network is trained and optimized by using the training data set, and the convolutional neural network algorithm enables the network to accurately identify and classify different parts through iterative learning and adjustment of network parameters. And evaluating the performance and accuracy of the model by using the verification set or the test set, and adjusting and optimizing the network structure or parameters according to the evaluation result so as to improve the performance of the model. The convolutional neural network algorithm applies the trained convolutional neural network model to an actual part library system. The parts are automatically identified and classified through model reasoning by inputting the part images to be identified, and are associated with the part information in the database.
By applying the convolutional neural network algorithm, the system can automatically identify and classify the parts in the input image, and correlate the parts with data in a part library through information such as image file names, classification, part numbers and the like, as shown in a table 1, which is a result of classifying and predicting the part image by applying the convolutional neural network algorithm;
TABLE 1 part image classification prediction results table
In the above embodiment, the part standardization module includes a parameter detection unit, a standard matching unit, and an error repair unit; the parameter detection unit reads and analyzes CAD files of the parts through CAD software, and compares whether the parts meet the requirements through a logic judgment algorithm; the standard matching unit compares the parameters and the characteristics of the parts with the matching degree of the standards through the matching rules, and when the parts are detected to be out of compliance with the standards, the error repairing unit automatically repairs the errors of the parts.
In a specific embodiment, the part standardization module detects whether parameters contained in the design part meet standardization requirements through the parameter detection unit. It will verify and compare parameters of the design part, such as dimensions, geometry, materials, etc. If the design parameters do not meet the standards or there is an abnormality, the parameter detection unit may prompt or warn so that the designer can adjust or correct. And matching the standard parts in the part library through a standard matching unit. The method can be compared and matched with standard parts in a part library according to the characteristics and the attributes of the designed parts. Through the standard matching unit, a designer can quickly find standardized parts that match design requirements, thereby avoiding the need for redesign or customization. The error repair unit helps the designer correct or repair errors or non-standard parts of the design. For example, when a parameter error is detected or a standard is not met, the error repair unit may provide an automatic or semi-automatic repair function to ensure that the designed part meets the standard and requirements.
Through the combined use of the parameter detection unit, the standard matching unit and the error repair unit, the part standardization module realizes the rapid and effective standardization detection and repair functions. The design method can help a designer to ensure that the designed parts meet the standard requirements, improve the consistency and reusability of the design, reduce errors and repeated work, and improve the design efficiency and the product quality.
In the above embodiment, the metadata management unit stores and organizes information and attribute data of the parts through a parts library database; the part library database stores unique identifiers, version numbers, attributes and CAD file path information of each part and comprises creation date, modification date, author, description and relation with other parts; the version control unit tracks and manages different versions of the parts through version management software SVN, and performs comparison and merging operations between the versions.
In a specific embodiment, the metadata management unit stores and manages metadata in the parts library by using a database. The database may provide structured data storage and query functions that may effectively organize and manage large amounts of metadata information. Meanwhile, the metadata management unit describes and organizes metadata by defining an appropriate data model, such as an Entity-Attribute-Relationship (EAR) model or other related data model. The data model may define attributes, relationships, constraints, etc. of the parts to provide efficient management and querying of metadata. In order to ensure consistency and interoperability of metadata, the metadata management unit needs to refer to and follow the relevant metadata standards. These criteria may include industry standards, international standards, or standards defined by specialized organizations for unifying naming, definition, and structure of metadata to facilitate sharing and integration of metadata. In addition, the metadata management unit provides the function of importing and updating metadata to ensure that the metadata in the parts library remains synchronized with the latest design or standard requirements. The designer may import new metadata into the parts library through the relevant interfaces or tools, and may modify existing metadata through update operations. The metadata management unit typically provides flexible retrieval and query functions that enable a user to quickly find and access desired metadata according to particular needs. This may include search functions based on keywords, attributes, classifications, or custom query conditions to provide efficient, accurate metadata query services.
In the above embodiment, the generation of the countermeasure network model generates the countermeasure samples by a generator and a discriminator to provide design choices and optimization suggestions, the generator generates the falsified samples by generating the model with random input as input data; the discriminator distinguishes the fake sample and the real sample generated by the generator through the discrimination model, the generation model and the discrimination model are trained alternately through an countermeasure training method, and network parameters are updated through a counter propagation algorithm, so that the generator is better in generating the real sample, and the discriminator is better in distinguishing the real sample and the fake sample; the method for generating the countermeasure network model countermeasure training comprises the following steps of;
step 1, randomly sampling from real data to serve as an input sample of a discriminator, and carrying out forward propagation of the discriminator to obtain a discrimination result of the discriminator on the real sample;
step 2, sampling from random input as an input sample of a generator, and performing forward propagation of the generator to generate a forged sample;
step 3, taking the sample generated by the generator and the real sample as the input of the discriminator, and carrying out forward propagation of the discriminator to obtain the discrimination result of the discriminator on the forged sample and the real sample;
Step 4, calculating a loss function of the discriminator according to a discrimination result of the discriminator, and carrying out back propagation to update parameters of the discriminator;
step 5, calculating a loss function of the generator according to a discrimination result obtained by a discriminator of the sample generated by the generator, and carrying out back propagation to update parameters of the generator;
and step 6, repeating the steps until the preset training iteration times or the convergence of the loss function are reached.
In a specific embodiment, as shown in fig. 2, the generation countermeasure network model generates dummy data by the generator, and the dummy data is discriminated together with the real data by the discriminator, and image data in the existing parts library is learned by the cyclic countermeasure between the generator and the discriminator to generate a new synthesized parts image. These generated images can be used to augment the contents of existing parts libraries, providing more alternative part samples. Based on the learned part data distribution, generating the countermeasure network model can generate part variants having similar styles but having differences. This can expand the variety of part libraries and provide alternative part choices for more different shapes, sizes, or features. At the same time, generating the countermeasure network model can be trained for detecting and repairing defects or errors in the part images. By learning the differences between the existing qualifying part image and the defective part image, the model may attempt to automatically identify and repair defects to improve part quality. In addition, generating the countermeasure network model may be used to enhance existing part images, such as to improve resolution, sharpness, or contrast of the image, etc. This helps to enhance the visualization of the part image, making it easier to understand and apply.
In the above embodiment, the model detection calculates the shape similarity between parts by a part shape similarity function based on a pre-constructed three-dimensional model and corresponding two-dimensional projection information, and verifies the parameter consistency thereof, the part shape similarity function expression is as follows,
S＝∑(α×A 1 +δ×A 2 +γ×A 3 ) 2 (3)
in the formula (3), S represents the shape similarity of the parts, A 1 、A 2 、A 3 Characteristic parameters of the part are represented, and alpha, delta and gamma are weight coefficients; the consistency of parameters is verified by comparing the shape similarity of different parts in the formula (3); if the shape similarity of the different parts is equal, the parameters are consistent;
the geometric analysis algorithm verifies whether the size parameters of different parts are consistent through the consistency of the size parameters; the dimensional parameter consistency is mathematically expressed as follows:
in the formula (4), P represents a dimensional parameter of the part, V represents an average value of the dimensional parameter, and ε is an allowable error threshold.
In a specific embodiment, the model detection uses a pre-built three-dimensional model and corresponding two-dimensional projection information as input data. The three-dimensional model is data describing the geometry and structure of the part, and the two-dimensional projection information is a projected representation of the part model on a two-dimensional plane. Meanwhile, model detection uses a shape similarity function to calculate shape similarity between parts based on a pre-built three-dimensional model and corresponding two-dimensional projection information. The shape similarity function can quantitatively evaluate the similarity degree between two parts according to different shape characteristics, topological structures and the like. In addition, model inspection also requires verification of parameter consistency between parts in the process of calculating shape similarity. The parameter consistency verification can judge whether the parameters of the components are consistent within a given standard range by comparing the parameters of the components, such as the attribute, the size, the material and the like. The results of the part inspection in the specific implementation are shown in table 2:
TABLE 2 part test results Table
In the data table 2, each part is subjected to model detection, and the results of shape similarity and parameter consistency are obtained through calculation. The value of shape similarity is between 0 and 1, 1 representing complete similarity, 0 representing complete dissimilarity; the parameter consistency indicates the consistency of the parameters of the parts and the standard, the consistency indicates that the parameters are in the standard range, and the inconsistency indicates that the parameters exceed the standard requirements.
In the above embodiment, the intelligent language model analyzes the user input and the design intent through a word splitting module, a part-of-speech tagging module, an entity recognition module and a syntax analysis module; the word segmentation module segments the text input by the user through the existing rule word segmentation method, and the word segmentation module is used for understanding sentence meaning; the part-of-speech tagging module predicts part-of-speech tags corresponding to words through a conditional random field model; the entity recognition module learns and retrieves the recognition word expression entity through a recurrent neural network and marks the recognition word expression entity in a classified manner; the syntactic analysis module analyzes the dependency relationship among the words through a dependency syntactic analyzer to infer the grammar structure of sentences and give analysis values, and the output end of the word segmentation module is connected with the input end of the part-of-speech tagging module; the output end of the part-of-speech tagging module is connected with the input end of the entity identification module; the output end of the entity identification module is connected with the input end of the syntactic analysis module.
In particular embodiments, the intelligent language model analyzes the input text and splits it into different words or words using natural language processing techniques through a word splitting module. Such a module helps to break down the incoming continuous text into smaller units of language, providing the basis for subsequent processing. The part-of-speech tagging module performs part-of-speech tagging on words obtained by segmentation, such as nouns, verbs, adjectives and the like, by using techniques such as language models, machine learning and the like. This module helps understand the grammatical and semantic functions of each word in context. The entity recognition module uses natural language processing and machine learning techniques to recognize and extract specific entities in the text, such as key information of part names, sizes, materials and the like. The module is used to identify and extract important entities associated with the part for further processing and analysis. The syntactic analysis module analyzes the structure and grammatical relation of the sentence using natural language processing and grammatical analysis techniques. The module is capable of identifying different components of subjects, predicates, objects, modifiers, etc., and identifying relationships between them. This helps understand the grammatical structure and semantic relationships of the input text.
In the implementation, the intelligent language model can effectively perform text processing tasks such as word segmentation, part-of-speech tagging, entity identification, syntactic analysis and the like. They help parse and understand the text information entered, thereby providing more accurate and intelligent language interaction functionality, further supporting the functionality and applications of the parts library system.
While specific embodiments of the present invention have been described above, it will be understood by those skilled in the art that these specific embodiments are by way of example only, and that various omissions, substitutions, and changes in the form and details of the methods and systems described above may be made by those skilled in the art without departing from the spirit and scope of the invention. For example, it is within the scope of the present invention to combine the above-described method steps to perform substantially the same function in substantially the same way to achieve substantially the same result. Accordingly, the scope of the invention is limited only by the following claims.
Claims (9)
1. A CAD software-based parts library system, the system comprising:
the CAD integration module integrates the part library system with CAD software through a CAD fusion plug-in unit so as to directly access and use components in the part library in the CAD software;
The part archiving module automatically identifies parts and rapidly classifies the parts in a grading way through the data collecting unit, the model training unit, the image identification classifying unit, the part information extracting unit and the automatic archiving unit;
the part standardization module automatically detects and verifies the parameter consistency of the parts in engineering design through a model detection and geometric analysis algorithm;
the intelligent search module is used for realizing an intelligent search function through an intelligent language model based on the part archiving module data; the intelligent language model automatically matches and recommends corresponding parts by analyzing keywords and descriptions input by a user, and provides associated classifications and labels;
the part version management module tracks and manages the change history and version information of the part through the metadata management unit and the version control unit and merges the part data;
the part community module establishes an online part sharing platform through a cloud server erection method so as to promote knowledge exchange and innovation cooperation among users;
the AI collaborative design module analyzes the design intent of the user through the intelligent language model and provides design choices and optimization suggestions through generating an countermeasure network model;
The output end of the part filing module is connected with the input end of the part standardization module, and the output end of the part standardization module is connected with the input end of the intelligent retrieval module; the output end of the intelligent search module is connected with the input end of the CAD integrated module, the output end of the part version management control module is connected with the input end of the part archiving module, and the output end of the AI collaborative design module is connected with the input end of the CAD integrated module.
2. A CAD software based parts library system as claimed in claim 1, wherein: the CAD integrated module comprises a data transmission unit, a user interface expansion unit, a CAD file processing unit and a synchronous updating unit; the data transmission unit acquires part metadata information from the part library system through an application program interface API; and transmitting the data to the CAD software via the application program interface API to access and use the part directly in the CAD; the user interface expansion unit adds a part library search column, a part preview image and a component selector in CAD software by introducing an interface element tool kit so as to facilitate browsing and selecting components in the part library; the CAD file processing unit realizes the importing and exporting functions of CAD file formats through a file format conversion tool so as to facilitate archiving management; when the components in the part library are changed, the CAD integration module updates the component list in the CAD software in real time through the synchronous updating unit, so that the user can always access the latest part information.
3. A CAD software based parts library system as claimed in claim 1, wherein: the data collection unit collects a large number of part images and corresponding label information through an internet retrieval tool, and a training data set is established; the model training unit trains the training data set through a convolutional neural network algorithm so as to learn the characteristics and the types of the parts; the image recognition and classification unit scans images through a scanner and automatically recognizes and classifies parts by using a trained model, part information is automatically extracted from the Internet through the part information extraction unit after recognition and classification are completed, and the part information extraction unit analyzes and extracts the acquired information through a data analysis tool; after the information extraction and analysis are completed, automatically archiving the parts subjected to identification and classification into a data model through an automatic archiving unit so as to facilitate searching and management; the automatic archiving unit records the categories and labels of the parts through a metadata recording tool to support the organization and retrieval of the files.
4. A CAD software based parts library system according to claim 3, wherein: the convolution neural network algorithm extracts features from an input image through a convolution function, a small area of the input image is given as a convolution kernel and a weight matrix, the convolution operation multiplies the convolution kernel and a local area of the image element by element, and the whole image is traversed through a window of the sliding convolution kernel; the expression of the convolution function F is as follows,
In the formula (1), M is one pixel point in the output feature map of the convolution operation, β is one pixel point of the input image, D is the weight of the convolution kernel, and g is the bias term; after the convolution is completed and the stacking is completed, the number of image pixel blocks is excessive, the size of the feature map is reduced through maximum pooling, the calculation complexity is reduced, and the main features are extracted; the maximum pooling function divides the input area into non-overlapping subareas, then selects the maximum value in each subarea as output, the formula of the maximum pooling function is that,
in the formula (2), b is a pixel point in an output characteristic diagram of the pooling operation, t is a sub-region in an input characteristic diagram, and θ is a pooling step size; multiple convolution layers, activation functions and pooling layers are repeatedly stacked in the maximum pooling to learn the details and abstract features of the input image and conduct accurate classification prediction.
5. A CAD software based parts library system as claimed in claim 1, wherein: the part standardization module comprises a parameter detection unit, a standard matching unit and an error repair unit; the parameter detection unit reads and analyzes CAD files of the parts through CAD software, and compares whether the parts meet the requirements through a logic judgment algorithm; the standard matching unit compares the parameters and the characteristics of the parts with the matching degree of the standards through the matching rules, and when the parts are detected to be out of compliance with the standards, the error repairing unit automatically repairs the errors of the parts.
6. A CAD software based parts library system as claimed in claim 5, wherein: the metadata management unit stores and organizes information and attribute data of the parts through a part library database; the part library database stores unique identifiers, version numbers, attributes and CAD file path information of each part and comprises creation date, modification date, author, description and relation with other parts; the version control unit tracks and manages different versions of the parts through version management software SVN, and performs comparison and merging operations between the versions.
7. A CAD software based parts library system as claimed in claim 1, wherein: the generation of the countermeasure network model generates countermeasure samples through a generator and a discriminator to provide design selection and optimization suggestions, and the generator generates forged samples by taking random input as input data through the generation model; the discriminator distinguishes the fake sample and the real sample generated by the generator through the discrimination model, the generation model and the discrimination model are trained alternately through an countermeasure training method, and network parameters are updated through a counter propagation algorithm, so that the generator is better in generating the real sample, and the discriminator is better in distinguishing the real sample and the fake sample; the method for generating the countermeasure network model countermeasure training comprises the following steps of;
Step 1, randomly sampling from real data to serve as an input sample of a discriminator, and carrying out forward propagation of the discriminator to obtain a discrimination result of the discriminator on the real sample;
step 2, sampling from random input as an input sample of a generator, and performing forward propagation of the generator to generate a forged sample;
step 3, taking the sample generated by the generator and the real sample as the input of the discriminator, and carrying out forward propagation of the discriminator to obtain the discrimination result of the discriminator on the forged sample and the real sample;
step 4, calculating a loss function of the discriminator according to a discrimination result of the discriminator, and carrying out back propagation to update parameters of the discriminator;
step 5, calculating a loss function of the generator according to a discrimination result obtained by a discriminator of the sample generated by the generator, and carrying out back propagation to update parameters of the generator;
and step 6, repeating the steps until the preset training iteration times or the convergence of the loss function are reached.
8. A CAD software based parts library system as claimed in claim 1, wherein: the model detection calculates the shape similarity between parts through a part shape similarity function based on a pre-constructed three-dimensional model and corresponding two-dimensional projection information, and verifies the parameter consistency, the part shape similarity function expression is as follows,
S＝∑(α×A 1 +δ×A 2 +γ×A 3 ) 2 (3)
In the formula (3), S represents the shape similarity of the parts, A 1 、A 2 、A 3 Characteristic parameters of the part are represented, and alpha, delta and gamma are weight coefficients; the consistency of parameters is verified by comparing the shape similarity of different parts in the formula (3); if the shape similarity of the different parts is equal, the parameters are consistent;
the geometric analysis algorithm verifies whether the size parameters of different parts are consistent through the consistency of the size parameters; the dimensional parameter consistency is mathematically expressed as follows:
in the formula (4), P represents a dimensional parameter of the part, V represents an average value of the dimensional parameter, and ε is an allowable error threshold.
9. A CAD software based parts library system as claimed in claim 1, wherein: the intelligent language model analyzes user input and design intention through a word splitting module, a part-of-speech labeling module, an entity recognition module and a syntactic analysis module; the word segmentation module segments the text input by the user through the existing rule word segmentation method, and the word segmentation module is used for understanding sentence meaning; the part-of-speech tagging module predicts part-of-speech tags corresponding to words through a conditional random field model; the entity recognition module learns and retrieves the recognition word expression entity through a recurrent neural network and marks the recognition word expression entity in a classified manner; the syntactic analysis module analyzes the dependency relationship among the words through a dependency syntactic analyzer to infer the grammar structure of sentences and give analysis values, and the output end of the word segmentation module is connected with the input end of the part-of-speech tagging module; the output end of the part-of-speech tagging module is connected with the input end of the entity identification module; the output end of the entity identification module is connected with the input end of the syntactic analysis module.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311002873.1A CN116976034A (en) | 2023-08-08 | 2023-08-08 | CAD software-based part library system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311002873.1A CN116976034A (en) | 2023-08-08 | 2023-08-08 | CAD software-based part library system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116976034A true CN116976034A (en) | 2023-10-31 |
Family
ID=88473017
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311002873.1A Pending CN116976034A (en) | 2023-08-08 | 2023-08-08 | CAD software-based part library system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116976034A (en) |
-
2023
- 2023-08-08 CN CN202311002873.1A patent/CN116976034A/en active Pending
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Soibelman et al. | Management and analysis of unstructured construction data types | |
| CN109783635A (en) | Use machine learning and fuzzy matching AUTOMATIC ZONING classifying documents and identification metadata | |
| US8346773B2 (en) | Product classification system | |
| US20210366055A1 (en) | Systems and methods for generating accurate transaction data and manipulation | |
| CN113987199B (en) | BIM intelligent image examination method, system and medium with standard automatic interpretation | |
| US20110060712A1 (en) | Method and system for design check knowledge construction | |
| US20230177267A1 (en) | Automated classification and interpretation of life science documents | |
| CN113486187A (en) | Buddhism knowledge graph construction method, device, equipment and storage medium | |
| CN113094512A (en) | Fault analysis system and method in industrial production and manufacturing | |
| CN112988982B (en) | Autonomous learning method and system for computer comparison space | |
| CN116721713B (en) | Data set construction method and device oriented to chemical structural formula identification | |
| CN116703337A (en) | Project document examination system and method based on artificial intelligence technology | |
| JP2007279978A (en) | Document retrieval device and document retrieval method | |
| CN116976034A (en) | CAD software-based part library system | |
| TWI793432B (en) | Document management method and system for engineering project | |
| Hast et al. | Making large collections of handwritten material easily accessible and searchable | |
| US20230134989A1 (en) | System and method for building document relationships and aggregates | |
| CN115640758B (en) | Three-dimensional model digital quality inspection method based on knowledge construction | |
| KR102331047B1 (en) | Method and System for creating and managing the relation based document | |
| EP4009194A1 (en) | Automated classification and interpretation of life science documents | |
| US20210295031A1 (en) | Automated classification and interpretation of life science documents | |
| Goyal et al. | Empowering Enterprise Architecture: Leveraging NLP for Time Efficiency and Strategic Alignment | |
| Koenig et al. | NEURAL-UML: Intelligent Recognition System of Structural Elements in UML Class Diagram | |
| EL KODSSI et al. | Applying Process Mining to Generate Business Process Models from Smart Environments | |
| ÓLAFSDÓTTIR | Using machine learning and natural language processing to automatically extract information from software documentation |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |