Assembly sequence generation method and system based on CATIA, electronic equipment and storage medium

Abstract

The application discloses a kind of based on CATIA's assembly sequence generation method, system, electronic equipment and storage medium, is related to industrial parts assembly technical field, the assembly sequence generation method based on CATIA includes: S10, the geometry and topological parameter of industrial parts is extracted, and assembly parameterization constraint is defined based on the assembly relationship between industrial parts;S20, sample data containing the geometry and topological parameter, the assembly parameterization constraint and corresponding artificial assembly operation sequence are collected, and the sample data is preprocessed and labeled, to construct the assembly dataset for model training;S30, assembly sequence prediction model is trained based on assembly dataset and loss function;S40, the assembly sequence prediction model predicts corresponding assembly operation sequence according to the parameterization information of input to-be-assembled industrial parts, and the assembly of industrial parts is automatically completed through CATIA platform.The beneficial effects of the application: improve assembly efficiency and assembly accuracy.

Description

Technical Field

[0001] This invention relates to the field of industrial parts assembly technology, and more specifically, to a CATIA-based assembly sequence generation method, system, electronic device, and storage medium. Background Technology

[0002] As the manufacturing industry accelerates its transformation towards intelligence and digitalization, computer-aided design software has become a core tool for product development. Among them, CATIA is a high-end integrated platform for computer-aided design, manufacturing, and analysis (CAD / CAM / CAE), widely used in complex product design fields such as aerospace, automotive manufacturing, and mechanical equipment. CATIA not only provides powerful 3D modeling capabilities but also supports multi-level assembly management from individual parts to entire systems, allowing designers to build digital product models by defining assembly constraints (such as fit, alignment, and concentricity).

[0003] However, traditional CATIA assembly operations rely heavily on manual interaction, and the assembly sequence of industrial parts needs to be manually compared, which is time-consuming and may contain errors, resulting in low assembly efficiency and accuracy. To this end, the present invention provides a CATIA-based assembly sequence generation method, system, electronic device, and storage medium that can predict the assembly sequence of industrial parts and automatically assemble the industrial parts according to the predicted assembly sequence, thereby improving assembly efficiency and assembly accuracy. Summary of the Invention

[0004] To overcome the shortcomings of existing technologies, this invention provides a CATIA-based assembly sequence generation method, system, electronic device, and storage medium that can predict the assembly sequence of industrial parts and automatically assemble the industrial parts according to the predicted assembly sequence, thereby improving assembly efficiency and assembly accuracy.

[0005] The technical solution adopted by this invention to solve its technical problem is: a CATIA-based assembly sequence generation method, the improvement of which is that the CATIA-based assembly sequence generation method includes the following steps: S10: Extract the geometric and topological parameters of industrial parts and define assembly parameterized constraints based on the assembly relationships between industrial parts. S20, collect sample data containing the geometric and topological parameters, the assembly parameterized constraints, and the corresponding manual assembly operation sequences, and preprocess and label the sample data to construct an assembly dataset for model training; S30, The assembly sequence prediction model is trained based on the assembly dataset and loss function; S40, the assembly sequence prediction model predicts the corresponding assembly operation sequence based on the parameterized information of the industrial parts to be assembled, and automatically completes the assembly of the industrial parts through the CATIA platform.

[0006] Furthermore, in step S10, the assembly parameterization constraints include axial constraints, surface constraints, axis constraints, and angle constraints.

[0007] Furthermore, in step S20, the specific method for preprocessing the sample data is as follows: the sample data is sequentially cleaned, normalized, PCA dimensionality reduced, and stored in JSON format.

[0008] Furthermore, in step S20, the assembly dataset includes a training set, a validation set, and a test set, and the training set, validation set, and test set are divided in a 7:2:1 ratio.

[0009] Furthermore, in step S30, the calculation expression for the loss function is: ; In the above formula, Represents the overall loss function. This represents the loss function for classification prediction based on assembly type. This represents the loss function predicted based on assembly constraint parameters; This represents the weighting coefficient, and its value ranges from 0 to 1.

[0010] Furthermore, the aforementioned The calculation expression is: ; In the above formula, This indicates the actual category label of the current industrial parts; The one-hot code representing the true category label of the current industrial part; This represents the model's predicted probability of the true class.

[0011] Furthermore, the aforementioned The calculation expression is: ; In the above formula, k represents the k parameters of the target assembly type. This represents the loss function based on the prediction of assembly constraint parameters when the assembly constraint parameters are linear parameters. This represents the loss function based on assembly constraint parameter prediction when the assembly constraint parameter is an angular parameter. Represents the true value of the linear parameter. This represents the predicted value of the linear parameter. This represents the true value of the angle parameter. This represents the predicted value of the angle parameter.

[0012] An assembly sequence generation system based on CATIA is applied to the CATIA-based assembly sequence generation method described above. The improvement lies in that the CATIA-based assembly sequence generation system includes: The assembly parameter extraction and constraint definition module is used to extract the geometric and topological parameters of industrial parts and define assembly parameterized constraints based on the assembly relationships between industrial parts. The data acquisition module is used to collect sample data including the geometric and topological parameters, the assembly parameterized constraints, and the corresponding manual assembly operation sequences; The data processing module is used to preprocess and label the sample data to construct an assembly dataset for model training. The model training module is used to train the assembly sequence prediction model using the assembly dataset and loss function; An assembly sequence prediction model is used to predict the corresponding assembly operation sequence based on the parameterized information of the input industrial parts to be assembled. The automated assembly module is used to automatically assemble industrial parts through the CATIA platform.

[0013] An electronic device, improved in that it includes at least one processor and at least one memory, wherein, The memory stores computer-readable instructions; The computer-readable instructions are executed by one or more processors, causing the electronic device to implement the CATIA-based assembly sequence generation method as described above.

[0014] A storage medium storing computer-readable instructions, wherein the improvement is that the computer-readable instructions are executed by one or more processors to implement the CATIA-based assembly sequence generation method as described above.

[0015] The beneficial effects of this invention are as follows: By extracting the geometric and topological parameters of industrial parts from the CATIA platform and defining assembly parameterization constraints based on the assembly relationships between industrial parts, this invention achieves the parameterization of industrial parts and assembly relationships, providing a standardized data foundation for the prediction and optimization of subsequent assembly sequences. By collecting sample data and preprocessing and labeling it, an assembly dataset for model training is constructed. The assembly sequence prediction model is trained based on the assembly dataset and a loss function to ensure that the model predicts the corresponding assembly operation sequence according to the parameterization information of the input industrial parts to be assembled, and automatically completes the assembly of the industrial parts through the CATIA platform. This replaces the traditional CATIA assembly operation, which heavily relies on manual interaction and requires manual comparison of the assembly sequences of industrial parts. It improves efficiency while avoiding potential errors in manual comparison, ensuring assembly accuracy. Therefore, this invention can predict the assembly sequence of industrial parts and automatically assemble them according to the predicted sequence, thereby improving assembly efficiency and accuracy. Attached Figure Description

[0016] Figure 1 This is a flowchart of an assembly sequence generation method based on CATIA according to the present invention; Figure 2 This is a block diagram of an assembly sequence generation system based on CATIA according to the present invention; Figure 3 This is a hardware structure diagram of an electronic device as an example embodiment; Figure 4 A block diagram illustrating an electronic device as an example embodiment; Figure 5 The flowchart of a CATIA-based assembly sequence generation method is shown as an example embodiment. Figure 1 . Detailed Implementation

[0017] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0018] The following will clearly and completely describe the concept, specific structure, and technical effects of the present invention in conjunction with embodiments and accompanying drawings, so as to fully understand the purpose, features, and effects of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, not all of them. Other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are all within the scope of protection of the present invention. Furthermore, all connections / linkages involved in the patent do not simply refer to direct contact between components, but rather to the ability to form a better connection structure by adding or reducing connecting accessories according to specific implementation conditions. The various technical features in this invention can be combined interactively without contradicting each other.

[0019] Reference Figure 1 and Figure 5 As shown, this invention discloses a CATIA-based assembly sequence generation method, which includes the following steps: S10: Extract the geometric and topological parameters of industrial parts and define assembly parameterized constraints based on the assembly relationships between industrial parts. It should be noted that, in this embodiment, the geometric and topological parameters of the industrial parts are defined as modifiable parameters that define the key dimensions (such as length and angle) and topological relationships (such as parallelism and concentricity) of the industrial parts. A driving relationship is established between these parameters and the geometric features of the 3D model. Multiple sets of parameter configurations are centrally managed through a predefined Excel spreadsheet, with each row representing a specific part specification (variant). When the user selects different configurations from the spreadsheet, CATIA automatically reads the corresponding parameter values ​​and updates the model via the ODBC interface, thereby quickly generating different part variants and achieving one-click creation and efficient management of serialized parts. The assembly parameterized constraints include union constraints, surface constraints, axis constraints, and angle constraints. These four types of assembly parameterized constraints can cover typical scenarios of industrial part assembly, enabling quantifiable description of assembly relationships. As a preferred embodiment, taking part P1 (the reference part) and part P2 (the constrained part) as objects, the parameterized description according to constraint type is as follows: Consistency constraints: Constraint identifier: C_Coincide-[serial number]; Core parameters: P1 feature (type: T1, ID: ID1), P2 feature (type: T2, ID: ID2), tolerance ε (ε∈[0,0.05]mm); Constraint equations: i, |gA_i-gB_i|≤ε (gA_i is the i-th geometric parameter of feature P1, and gB_i is the i-th geometric parameter of feature P2); Surface constraints (including mating constraints and parallel surface constraints): Fitting constraints: Constraint identifier: C_Face-Contact-[serial number]; Core parameters: P1 plane (equation: a1x+b1y+c1z+d1=0), P2 plane (equation: a2x+b2y+c2z+d2=0); Constraints: Normal vectors are reversed ((a1,b1,c1)=-k(a2,b2,c2), k>0), planar distance=0; Parallel plane constraint: Constraint identifier: C_Face-Parallel-[serial number]; Core parameters: P1 plane (equation: a1x+b1y+c1z+d1=0), P2 plane (equation: a2x+b2y+c2z+d2=0), distance d (d≥0mm); Constraints: Normal vectors are in the same direction ((a1,b1,c1)=k(a2,b2,c2), k>0), Plane distance = d; Axis constraints (including coaxial constraints and parallel axis constraints): Coaxial constraint: Constraint identifier: C_Axis-Coaxial-[serial number]; Core parameters: P1 axis (point-to-point: (x-x1) / u1=(y-y1) / v1=(z-z1) / w1), P2 axis (point-to-point: (x-x2) / u2=(y-y2) / v2=(z-z2) / w2); Constraints: Direction vectors are in the same direction ((u1,v1,w1) = k (u2,v2,w2), k>0), axis distance = 0; Parallel axis constraint: Constraint identifier: C_Axis-Parallel-[serial number]; Core parameters: P1 axis (point-to-point: (x-x1) / u1=(y-y1) / v1=(z-z1) / w1), P2 axis (point-to-point: (x-x2) / u2=(y-y2) / v2=(z-z2) / w2), distance d (d≥0mm); Constraints: Direction vectors are in the same direction ((u1,v1,w1) = k(u2,v2,w2), k>0), axis distance = d; Angle constraints: Constraint identifier: C_Angle-[serial number]; Core parameters: P1 feature vector V1=(u1,v1,w1), P2 feature vector V2=(u2,v2,w2), target angle θ (0°<θ<180°), tolerance Δθ=±0.5°; Constraint equation: θ - Δθ ≤ arccos[(V1 V2) / (|V1| |V2|)]≤θ+Δθ; Through the above parameterized expression, assembly constraints are transformed into quantifiable and computable digital models, realizing a digital description of assembly relationships. Moreover, this expression system is compatible with CATIA's constraint solver, supports automatic detection and resolution of constraint conflicts, and provides a standardized data foundation for the prediction and optimization of subsequent assembly sequences. In addition, in the above formula, k represents a positive proportionality coefficient, which represents the length ratio of the two direction vectors.

[0020] S20, collect sample data containing the geometric and topological parameters, the assembly parameterized constraints, and the corresponding manual assembly operation sequences, and preprocess and label the sample data to construct an assembly dataset for model training; specifically, the sample data is preprocessed in the following ways: data cleaning, normalization, PCA dimensionality reduction, and storage in JSON format are performed sequentially; in addition, the assembly dataset includes a training set, a validation set, and a test set, and the training set, validation set, and test set are divided in a 7:2:1 ratio.

[0021] It should be noted that, in this embodiment, the construction of the assembly dataset follows a "collection-preprocessing-annotation-verification" process: collecting historical cases and virtual data generated by CATIA scripts; preprocessing includes data cleaning, min-max normalization, and PCA dimensionality reduction (retaining 95% of features); annotation adopts an "automatic + manual verification" mode. Furthermore, the sample data collection needs to obtain data from multiple sources, including actual assembly cases, virtual assembly experiments, and existing assembly databases. These data cover various types of parts, assembly scenarios, and assembly processes, ensuring the diversity and representativeness of the dataset. Moreover, by collecting actual assembly cases, real assembly process data can be obtained, including the assembly sequence of parts, assembly... Information such as assembly method and assembly time reflects the actual assembly situation in production, providing a reliable basis for model learning. During data preprocessing, data cleaning uses the Laida criterion (3σ criterion) to remove outliers, such as samples with constraint parameters exceeding the reasonable range. Data normalization uses the min-max standardization method to map parameter values ​​to the [0,1] interval, eliminating the influence of dimensions. Furthermore, the preprocessed data is stored in JSON format, establishing an association index of "part parameters - constraint parameters - assembly sequence". Finally, the preprocessed sample data totals nearly 100,000, with annotations including operation type, object, and parameters. The sample data is divided into training / validation / test sets in a 7:2:1 ratio, providing high-quality sample support for the model.

[0022] S30, The assembly sequence prediction model is trained based on the assembly dataset and loss function; It should be noted that, in this embodiment, the assembly sequence prediction model needs to design a hybrid loss function for the joint test task of "assembly type classification and constraint parameter regression," which includes a loss function based on assembly type classification prediction and a loss function based on assembly constraint parameter prediction. The loss function based on assembly type classification prediction measures the difference between the model's predicted assembly type probability distribution and the true distribution, guiding the model to make correct classifications, thereby optimizing the model's discrete judgment ability and ensuring the accuracy of assembly type identification. The loss function based on assembly constraint parameter prediction measures the error between the model's predicted constraint parameter values ​​and the true values, guiding the model to make accurate fits, thereby optimizing the model's continuous numerical prediction ability and ensuring the accuracy of constraint parameter estimation. Specifically, the calculation expression of the loss function is as follows: ; In the above formula, Represents the overall loss function. This represents the loss function for classification prediction based on assembly type. This represents the loss function predicted based on assembly constraint parameters; This represents the weighting coefficient, and its value ranges from 0 to 1. Furthermore, the aforementioned The calculation expression is: ; In the above formula, This indicates the actual category label of the current industrial parts; c represents the one-hot encoding of the current industrial part's true category label; c represents the current industrial part's true category index. This represents the model's predicted probability of the true class. Used to measure whether the probability of the model assigning the true assembly type is large enough; when the model predicts the probability of the true class... The closer it gets to 1, The closer to 0, the more accurate the classification; As a preferred embodiment, the Furthermore, the aforementioned one-hot encoding is used to transform the true category label of the current industrial part into a vector that can be directly mathematically operated on with the same probability distribution. Specifically, the model output is a predicted probability distribution obtained through Softmax. , The calculation expression is: , ; In the above formula, z represents the original output of the model; m represents the index of each category of industrial parts, used to traverse the category index of industrial parts; This represents the softmax function, a core tool in multi-class classification tasks in machine learning, primarily used to transform a set of arbitrary real numbers (the original output z of the model in this embodiment) into a probability distribution; The The calculation expression is: ; In the above formula, n represents the n parameters of the target assembly type. This represents the loss function based on the prediction of assembly constraint parameters when the assembly constraint parameters are linear parameters. This represents the loss function based on assembly constraint parameter prediction when the assembly constraint parameter is an angular parameter. Represents the true value of the linear parameter. This represents the predicted value of the linear parameter. This represents the true value of the angle parameter. This represents the predicted value of the angle parameter; Furthermore, The calculation expression is: ; The calculation expression is: ; In the above formula, By taking the "minimum angular error in the sense of period," a sudden change in loss is avoided when the angle spans 0° / 360°. As a preferred embodiment, when... , At that time, , Therefore, the error is calculated as 7° instead of 353°.

[0023] It should be noted that, in this embodiment, the assembly sequence prediction model adopts a Self-AttentionDecoder architecture, including an embedding layer, four Self-Attention layers, an FFN, and an output layer. It captures long-distance dependencies between operations through Self-Attention to solve the problem of multi-constraint coupling. Furthermore, during the assembly sequence prediction training process, after processing by multiple self-attention layers, the assembly sequence prediction model can fully learn the complex patterns and rules in the assembly operation sequence. The output of the self-attention layer is mapped to the specific assembly operation space through a fully connected layer, and the predicted assembly operation sequence is output. Each operation sequence then predicts classification constraint parameters. That is, the user inputs a set of parts, and the model outputs the assembly type and the corresponding constraint parameters in sequence. S40, the assembly sequence prediction model predicts the corresponding assembly operation sequence based on the parameterized information of the industrial parts to be assembled, and automatically completes the assembly of the industrial parts through the CATIA platform.

[0024] It should be noted that, in this embodiment, the assembly sequence prediction model establishes a mapping relationship between the predicted assembly operation sequence and the CATIA operation interface, and calls the CATIA interface based on the mapping relationship to convert the predicted assembly operations into actual assembly actions. Error detection and processing are performed during the assembly process to achieve automatic assembly of parts.

[0025] Reference Figure 2 As shown, the present invention also discloses a CATIA-based assembly sequence generation system 600, applied to a CATIA-based assembly sequence generation method as described in the above embodiments. The CATIA-based assembly sequence generation system includes: Assembly parameter extraction and constraint definition module 601 is used to extract the geometric and topological parameters of industrial parts and define assembly parameterized constraints based on the assembly relationship between industrial parts. The data acquisition module 602 is used to acquire sample data including the geometric and topological parameters, the assembly parameterized constraints, and the corresponding manual assembly operation sequence; The data processing module 603 is used to preprocess and label the sample data to construct an assembly dataset for model training. The model training module 604 is used to train the assembly sequence prediction model using the assembly dataset and the loss function; Assembly sequence prediction model 605 is used to predict the corresponding assembly operation sequence based on the parameterized information of the input industrial parts to be assembled. The Automated Assembly Module 606 is used to automatically assemble industrial parts through the CATIA platform.

[0026] It should be noted that the CATIA-based assembly sequence generation strategy provided in the above embodiments is only an example of the division of the above functional modules. In practical applications, the above functions can be assigned to different functional modules as needed. That is, the internal structure of the CATIA-based assembly sequence generation system 600 will be divided into different functional modules to complete all or part of the functions described above. Furthermore, the CATIA-based assembly sequence generation system 600 and the CATIA-based assembly sequence generation method provided in the above embodiments belong to the same concept. The specific way in which each module performs operations has been described in detail in the method embodiments, and will not be repeated here.

[0027] Figure 3 A schematic diagram of the structure of an electronic device according to an exemplary embodiment is shown.

[0028] It should be noted that this electronic device is merely an example adapted to the present invention and should not be construed as providing any limitation on the scope of the present invention. Furthermore, this electronic device should not be interpreted as requiring or depending on any particular feature. Figure 3 One or more components of the exemplary electronic device 2000 shown.

[0029] The hardware structure of electronic devices 2000 can vary significantly due to differences in configuration or performance, such as... Figure 3 As shown, the electronic device 2000 includes: a power supply 210, an interface 230, at least one memory 250, and at least one central processing unit (CPU) 270.

[0030] Specifically, power supply 210 is used to provide operating voltage for various hardware devices on electronic device 2000.

[0031] Interface 230 includes at least one wired or wireless network interface 231 for interacting with external devices. Of course, in other examples adapted to this invention, interface 230 may further include at least one serial-to-parallel conversion interface 233, at least one input / output interface 235, and at least one USB interface 237, etc. Figure 3 As shown, this does not constitute a specific limitation.

[0032] The memory 250 serves as a carrier for resource storage and can be a read-only memory, random access memory, disk, or optical disk, etc. The resources stored on it include the operating system 251, application programs 253, and data 255, etc., and the storage method can be temporary storage or permanent storage.

[0033] The operating system 251 is used to manage and control the various hardware devices and application programs 253 on the electronic device 2000, so as to enable the central processing unit 270 to perform calculations and processing on the massive data 255 in the memory 250. It can be Windows Server™, Mac OS X™, Unix™, Linux™, FreeBSD™, etc.

[0034] Application 253 is a computer-readable instruction based on operating system 251 that performs at least one specific task, and may include at least one module ( Figure 3 (Not shown), each module may contain computer-readable instructions for the electronic device 2000. For example, a device for prioritizing tasks may be considered as an application program 253 deployed on the electronic device 2000.

[0035] Data 255 may be signal information, etc., and is stored in memory 250.

[0036] The central processing unit 270 may include one or more processors and is configured to communicate with the memory 250 via at least one communication bus to read computer-readable instructions stored in the memory 250, thereby performing operations and processing on massive amounts of data 255 stored in the memory 250. For example, a CATIA-based assembly sequence generation method can be implemented by the central processing unit 270 reading a series of computer-readable instructions stored in the memory 250.

[0037] Furthermore, the present invention can also be implemented through hardware circuits or a combination of hardware circuits and software. Therefore, the implementation of the present invention is not limited to any specific hardware circuit, software, or combination thereof.

[0038] Please see Figure 4 This invention provides an electronic device 4000, which may include: a desktop computer, a laptop computer, a server, etc., with sensor recognition capabilities.

[0039] exist Figure 4 In this context, the electronic device 4000 includes at least one processor 4001 and at least one memory 4003.

[0040] The data interaction between the processor 4001 and the memory 4003 can be achieved through at least one communication bus 4002. This communication bus 4002 may include a path for transmitting data between the processor 4001 and the memory 4003. The communication bus 4002 may be a PCI (Peripheral Component Interconnect) bus or an EISA (Extended Industry Standard Architecture) bus, etc. The communication bus 4002 can be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 4 The bus is represented by a single thick line, but this does not mean that there is only one bus or one type of bus.

[0041] Optionally, the electronic device 4000 may further include a transceiver 4004, which can be used for data interaction between the electronic device and other electronic devices, such as sending and / or receiving data. It should be noted that in practical applications, the transceiver 4004 is not limited to one type, and the structure of the electronic device 4000 does not constitute a limitation on the embodiments of the present invention.

[0042] Processor 4001 may be a CPU (Central Processing Unit), a general-purpose processor, a DSP (Digital Signal Processor), an ASIC (Application Specific Integrated Circuit), an FPGA (Field Programmable Gate Array), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. It can implement or execute the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this invention. Processor 4001 may also be a combination that implements computational functions, such as including one or more microprocessor combinations, a combination of a DSP and a microprocessor, etc.

[0043] The memory 4003 may be a ROM (Read Only Memory) or other type of static storage device capable of storing static information and instructions, RAM (Random Access Memory) or other type of dynamic storage device capable of storing information and instructions, or an EEPROM (Electrically Erasable Programmable Read Only Memory), CD-ROM (Compact Disc Read Only Memory) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disk storage media or other magnetic storage devices, or any other medium capable of carrying or storing desired program instructions or code in the form of instructions or data structures and accessible by the electronic device 4000, but not limited thereto.

[0044] The memory 4003 stores computer-readable instructions, and the processor 4001 can read the computer-readable instructions stored in the memory 4003 through the communication bus 4002.

[0045] The computer-readable instructions are executed by one or more processors 4001 to implement the CATIA-based assembly sequence generation method in the above embodiments.

[0046] Furthermore, this embodiment of the invention provides a storage medium storing computer-readable instructions, which are executed by one or more processors to implement the CATIA-based assembly sequence generation method described above.

[0047] Compared with the prior art, the beneficial effects of the present invention are as follows: (1) The assembly relationship is abstracted into quantifiable parameters and constraints to achieve digital description, which solves the problems of low efficiency and poor consistency of traditional experience design, shortens the design cycle and improves accuracy; (2) The assembly dataset construction module is the guarantee of model performance. It ensures diversity by collecting multi-source data, improves quality through 3σ cleaning, normalization and other preprocessing, and ensures accuracy through "automatic + manual" annotation, providing high-quality samples for the model and further optimizing assembly accuracy. (3) Introducing Attention to solve the assembly serialization problem and mining the relationship between the assembly process can help predict the assembly sequence and parameters of parts in batches, thereby further optimizing assembly efficiency.

[0048] The above is a detailed description of the preferred embodiments of the present invention. However, the present invention is not limited to the embodiments described. Those skilled in the art can make various equivalent modifications or substitutions without departing from the spirit of the present invention. All such equivalent modifications or substitutions are included within the scope defined by the claims of this application.

Claims

1. A method for generating assembly sequences based on CATIA, characterized in that, The CATIA-based assembly sequence generation method includes the following steps: S10: Extract the geometric and topological parameters of industrial parts and define assembly parameterized constraints based on the assembly relationships between industrial parts. S20, collect sample data containing the geometric and topological parameters, the assembly parameterized constraints, and the corresponding manual assembly operation sequences, and preprocess and label the sample data to construct an assembly dataset for model training; S30, The assembly sequence prediction model is trained based on the assembly dataset and loss function; S40, the assembly sequence prediction model predicts the corresponding assembly operation sequence based on the parameterized information of the industrial parts to be assembled, and automatically completes the assembly of the industrial parts through the CATIA platform.

2. The assembly sequence generation method based on CATIA according to claim 1, characterized in that, In step S10, the assembly parameterization constraints include axial constraints, surface constraints, axis constraints, and angle constraints.

3. The assembly sequence generation method based on CATIA according to claim 1, characterized in that, In step S20, the specific method for preprocessing the sample data is as follows: the sample data is sequentially cleaned, normalized, PCA dimensionality reduced, and stored in JSON format.

4. The assembly sequence generation method based on CATIA according to claim 1, characterized in that, In step S20, the assembly dataset includes a training set, a validation set, and a test set, and the training set, validation set, and test set are divided in a 7:2:1 ratio.

5. The assembly sequence generation method based on CATIA according to claim 1, characterized in that, In step S30, the calculation expression for the loss function is: ； In the above formula, Represents the overall loss function. This represents the loss function for classification prediction based on assembly type. This represents the loss function predicted based on assembly constraint parameters; This represents the weighting coefficient, and its value ranges from 0 to 1.

6. The assembly sequence generation method based on CATIA according to claim 5, characterized in that, The The calculation expression is: ； In the above formula, This indicates the actual category label of the current industrial parts; The one-hot code representing the true category label of the current industrial part; This represents the model's predicted probability of the true class.

7. The assembly sequence generation method based on CATIA according to claim 5, characterized in that, The The calculation expression is: ； In the above formula, k represents the k parameters of the target assembly type. This represents the loss function based on the prediction of assembly constraint parameters when the assembly constraint parameters are linear parameters. This represents the loss function based on assembly constraint parameter prediction when the assembly constraint parameter is an angular parameter. Represents the true value of the linear parameter. This represents the predicted value of the linear parameter. This represents the true value of the angle parameter. This represents the predicted value of the angle parameter.

8. A CATIA-based assembly sequence generation system, applied to the CATIA-based assembly sequence generation method according to any one of claims 1-7, characterized in that, The CATIA-based assembly sequence generation system includes: The assembly parameter extraction and constraint definition module is used to extract the geometric and topological parameters of industrial parts and define assembly parameterized constraints based on the assembly relationships between industrial parts. The data acquisition module is used to collect sample data including the geometric and topological parameters, the assembly parameterized constraints, and the corresponding manual assembly operation sequences; The data processing module is used to preprocess and label the sample data to construct an assembly dataset for model training. The model training module is used to train the assembly sequence prediction model using the assembly dataset and loss function; An assembly sequence prediction model is used to predict the corresponding assembly operation sequence based on the parameterized information of the input industrial parts to be assembled. The automated assembly module is used to automatically assemble industrial parts through the CATIA platform.

9. An electronic device, characterized in that, Includes at least one processor and at least one memory, wherein, The memory stores computer-readable instructions; The computer-readable instructions are executed by one or more processors, causing the electronic device to implement the CATIA-based assembly sequence generation method as described in any one of claims 1-7.

10. A storage medium having computer-readable instructions stored thereon, characterized in that, The computer-readable instructions are executed by one or more processors to implement the CATIA-based assembly sequence generation method according to any one of claims 1-7.

Images