System and method for detecting the fit of a manufactured piece

Abstract

The application provides a manufacturing piece fitting detection system and a fitting detection method. The fitting detection system comprises a determination module configured to determine a plurality of to-be-detected parts in a manufacturing piece based on detection requirements; and an artificial intelligence model configured to determine a connection type between the plurality of to-be-detected parts based on the determined plurality of to-be-detected parts. The artificial intelligence model is further configured to detect a fitting relationship between the plurality of to-be-detected parts and determine whether the fitting relationship meets a design specification of the manufacturing piece according to the connection type. The fitting detection system provided by the application first determines the connection type between the to-be-detected parts by using the artificial intelligence model, and then determines whether the fitting relationship meets the requirements based on the connection type and the design specification, so that the reasonable fitting relationship in design can be avoided from being misjudged as a design problem, the accuracy of detection and judgment is improved, and the artificial intelligence model is used to automatically check whether each fitting relationship of the to-be-detected parts is reasonable, so that the detection accuracy and efficiency of the fitting relationship of the manufacturing piece are improved.

Description

Technical Field

[0001] This invention relates to the field of machinery, and more specifically, to a system for detecting the fit of manufactured parts, a method for detecting the fit, a computer device for implementing the method for detecting the fit, a computer-readable storage medium for executing the method for detecting the fit, and a computer program product for implementing the method for detecting the fit. Background Technology

[0002] In the current machinery manufacturing industry, 3D design software (such as CATIA, Solidworks, etc.) is usually used in the design stage of manufactured parts to simulate and evaluate the fit relationship between the parts, such as clearance and interference, so as to discover potential problems in the design stage and avoid errors and cost waste in the subsequent production and assembly process.

[0003] However, current 3D design software primarily relies on geometric judgment when detecting fit relationships. This involves analyzing the geometry, dimensions, and positional relationships between parts to determine the presence of gaps or interferences. However, in mechanical manufacturing, certain gaps and interferences may be necessary for the design to meet specific functional or performance requirements. Current 3D design software, when detecting fit relationships, cannot identify these fit relationships that meet specific design requirements; instead, it treats them as design problems.

[0004] Consequently, a large number of false problems will be reported during the testing period—that is, problems that are actually design requirements rather than actual issues. This not only increases the workload of designers but may also mislead them in their judgment and handling of real problems. Summary of the Invention

[0005] The purpose of this invention is to solve the problems existing in the prior art and provide a system and method for detecting the fit of manufactured parts, so as to improve the accuracy and efficiency of detecting the fit relationship of manufactured parts in a cost-effective and reliable manner.

[0006] Therefore, according to one aspect of the present invention, a mating inspection system for manufactured parts is provided, the mating inspection system comprising: a determination module configured to determine a plurality of parts to be inspected in the manufactured part based on inspection requirements; and an artificial intelligence model configured to determine the connection type between the plurality of parts to be inspected based on the determined plurality of parts to be inspected, wherein the artificial intelligence model is further configured to: detect the mating relationship between the plurality of parts to be inspected, and determine whether the mating relationship conforms to the design specifications of the manufactured part according to the connection type.

[0007] Based on the above-described technical concept, the present invention may further include any one or more of the following optional forms.

[0008] In some alternative forms, the artificial intelligence model is a trained artificial intelligence model, and the coordination detection system includes a training module configured to train the artificial intelligence model using training data, wherein the training data includes text data and / or three-dimensional data.

[0009] In some alternative forms, the text data includes at least a portion of the part name and / or the region where the part is located and / or the part shape and / or the part connection type and / or the design specifications of the manufactured part, and the three-dimensional data includes a three-dimensional model of the part and / or the part assembly relationship and / or the part mechanical or simulation data.

[0010] In some alternative configurations, the determining module is configured to determine the plurality of parts to be inspected as primary inspection parts and environmental parts.

[0011] In some alternative configurations, the determining module is configured to determine the main inspection component based on inspection requirements, and to determine the component to be inspected within a preset distance from the main inspection component as the environmental component.

[0012] In some alternative forms, the determining module is configured to determine the main inspection part and the environmental part based on a parts list, wherein the parts list is pre-stored in the determining module.

[0013] In some alternative forms, the determining module is configured to partition the plurality of parts to be inspected according to the determined boundaries of the main inspection part and the environmental part; the artificial intelligence model is configured to determine the connection type between the main inspection part and the environmental part in each part's region, detect the mating relationship between the main inspection part and the environmental part in each part's region, and determine whether the mating relationship between the main inspection part and the environmental part in each part's region conforms to the design specifications of the manufactured part based on the connection type.

[0014] In some alternative forms, the mating detection system includes a storage module configured to store at least a portion of the design specifications of the manufactured part, and / or the storage module is integrated with the artificial intelligence model.

[0015] In some alternative forms, the mating inspection system includes a notification module configured to send an improvement notification to a user if the mating relationship between the plurality of parts to be inspected does not conform to the design specifications of the manufactured part, so that the user can improve the mating relationship.

[0016] In some alternative forms, the notification module is deployed in the cloud and / or at the edge and / or on a terminal, and is integrated into 3D design software to send improvement notifications to users through the 3D design software.

[0017] In some alternative forms, the determining module is integrated into the artificial intelligence model, or the determining module is integrated into the 3D design software.

[0018] In some alternative forms, the artificial intelligence model and / or the determining module are deployed in the cloud and / or at the edge and / or on a terminal.

[0019] In some alternative forms, the artificial intelligence model is a generative artificial intelligence model, which includes a semi-supervised self-learning model.

[0020] In some alternative forms, the artificial intelligence model is integrated into 3D design software for matching detection via the 3D design software.

[0021] In some alternative forms, the mating relationship includes clearance fit and / or interference fit, and / or the connection type includes snap-fit ​​and / or welding and / or threaded connection and / or riveting.

[0022] According to another aspect of the present invention, a method for mating inspection of a manufactured part is provided, the method comprising: determining a plurality of parts to be inspected in the manufactured part based on inspection requirements; determining the connection type between the plurality of parts to be inspected using an artificial intelligence model based on the determined plurality of parts to be inspected; detecting the mating relationship between the plurality of parts to be inspected using an artificial intelligence model; and determining whether the mating relationship conforms to the design specifications of the manufactured part based on the connection type using an artificial intelligence model.

[0023] In some alternative forms, the matching detection method includes training the artificial intelligence model using training data.

[0024] In some alternative forms, identifying a plurality of parts to be inspected in the manufactured part includes: identifying the plurality of parts to be inspected as primary inspection parts and environmental parts.

[0025] In some alternative forms, determining multiple parts to be inspected in the manufactured part includes: determining the main inspection part based on inspection requirements, and determining the parts to be inspected within a preset distance from the main inspection part as the environmental parts.

[0026] In some alternative forms, determining multiple parts to be inspected in the manufactured part includes: determining the main inspection part and the environmental parts based on a pre-stored parts list.

[0027] In some optional forms, the mating detection method includes: dividing the plurality of parts to be detected into zones based on the determined boundaries of the main detection part and the environmental parts; determining the connection type between the main detection part and the environmental parts in each part's zone using an artificial intelligence model; detecting the mating relationship between the main detection part and the environmental parts in each part's zone using the artificial intelligence model; and determining, based on the connection type, whether the mating relationship between the main detection part and the environmental parts in each part's zone conforms to the design specifications of the manufactured part using the artificial intelligence model.

[0028] In some alternative forms, the fit detection method includes sending an improvement notification to the user when the fit relationship between the plurality of parts to be tested does not conform to the design specifications of the manufactured part, so that the user can improve the fit relationship.

[0029] According to another aspect of the present invention, a computer device is provided, the computer device including a memory, a processor and instructions stored in the memory and executable by the processor, wherein the processor executes the instructions to implement the above-described method for detecting the fit of a manufactured part.

[0030] According to another aspect of the present invention, a computer-readable storage medium is provided having computer-executable instructions stored thereon for performing the above-described method for detecting the fit of a manufactured part.

[0031] According to another aspect of the present invention, a computer program product is provided, comprising computer-executable instructions that, when executed by at least one processor, implement the above-described method for detecting the fit of a manufactured part.

[0032] The mating inspection system and method for manufactured parts provided by this invention first determine the connection type between the parts to be inspected using an artificial intelligence model, and then determine whether the mating relationship meets the requirements based on the connection type and design specifications. This can avoid misjudging a reasonable mating relationship as a design problem, improve the accuracy of inspection and judgment, and eliminate the need for manual inspection of each mating relationship of the parts to be inspected, thereby improving the accuracy and efficiency of the inspection of the mating relationship of manufactured parts and enhancing the efficiency of design and manufacturing. Attached Figure Description

[0033] Other features and advantages of the present invention will be better understood through the following detailed description of optional embodiments in conjunction with the accompanying drawings, wherein:

[0034] Figure 1 A schematic diagram of a manufacturing part fitting detection system according to an embodiment of the present invention is shown;

[0035] Figure 2 It shows that it is suitable for passing through Figure 1 The rear bumper of the vehicle being inspected using the detection system.

[0036] Figure 3 It shows Figure 2 The rear bumper is divided into sections;

[0037] Figure 4 A schematic flowchart of a method for testing the fit of manufactured parts according to an embodiment of the present invention is shown;

[0038] Figure 5 A schematic flowchart of a method for testing the fit of manufactured parts according to another embodiment of the present invention is shown;

[0039] Figure 6 A schematic flowchart of a method for testing the fit of manufactured parts according to another embodiment of the present invention is shown;

[0040] Figure 7 A schematic flowchart of a method for detecting the fit of manufactured parts according to another embodiment of the present invention is shown; and

[0041] Figure 8 A schematic diagram of a computer device according to an embodiment of the present invention is shown. Detailed Implementation

[0042] The implementation and use of the embodiments are discussed in detail below. While the exemplary methods and systems described below include software and / or firmware executed on hardware within other components, it should be noted that these examples are merely illustrative and should not be considered limiting. Therefore, although exemplary methods and systems have been described below, those skilled in the art will readily understand that the specific embodiments discussed are merely exemplary of particular ways of implementing and using the invention, and not intended to limit the scope of the invention.

[0043] Furthermore, the flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of methods and systems according to various embodiments of the present invention. It should be noted that the functions indicated in the blocks may occur in a different order than that shown in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, or they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the flowcharts and / or block diagrams, and combinations of blocks in the flowcharts and / or block diagrams, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0044] In modern manufacturing, to ensure product assembly quality and design accuracy, it is typically necessary to inspect the mating relationships between manufactured parts during the design phase. Currently, 3D design software is used to determine whether these mating relationships are erroneous. The inventors have discovered that because 3D design software primarily relies on geometric calculations to make geometric judgments about the mating relationships, it cannot identify the design intent. For example, in transmission mechanisms, interference between gears can be used to improve transmission stability and efficiency, or in some manufactured parts, reserved clearances can ensure normal operation during thermal expansion and contraction. However, these design requirements are often not directly understood by 3D design software, which treats all interferences and clearances as design problems during inspection. Therefore, existing inspection methods may prolong product development cycles, requiring designers to invest more time and effort in troubleshooting, and potentially leading to mating errors in subsequent production stages, thus affecting product performance and manufacturing efficiency. Therefore, improvements are needed to the inspection of the mating relationships of manufactured parts to enhance the accuracy and efficiency of the inspection process.

[0045] Reference Figure 1A manufacturing part fitting inspection system 100 according to one embodiment of the present invention mainly includes a determination module 110 and an artificial intelligence model 120. The determination module 110 is used to determine multiple parts to be inspected in the manufactured part based on inspection requirements. In some embodiments, the inspection requirements can be determined according to the function or importance of the parts (e.g., critical functional parts, high-risk parts, core parts, etc.), and / or according to the manufacturing process of the parts (e.g., complex process parts, special process parts, etc.), and / or according to the historical performance of the parts (e.g., rework parts, first-time application parts, etc.). It should be understood that the method of determining the inspection requirements is not limited to this and can be changed as needed. The artificial intelligence model 120 determines the connection type between the multiple parts to be inspected based on the determined multiple parts to be inspected. Specifically, the artificial intelligence model 120 can determine the connection type between the parts to be inspected by recognizing the characteristics of the parts to be inspected (e.g., part name, part shape, etc.). In some embodiments, the connection type can include snap-fit, welding, threaded connection, riveting, etc. For example, when the manufactured parts are vehicle components, multiple parts to be inspected may include a rear bumper and a rear bumper trim. If the AI ​​model 120 identifies multiple parts as rear bumpers and rear bumper trims by their names, it will determine that the connection type is a snap-fit. Furthermore, the AI ​​model 120 also detects the mating relationships between the multiple parts and determines whether these relationships conform to the design specifications of the manufactured parts based on the connection type. For instance, after the AI ​​model 120 determines that the connection type between the rear bumper and the rear bumper trim is a snap-fit, it will then detect the mating relationship between them. In this example, the rear bumper and the rear bumper trim use a clearance fit. Subsequently, if the connection type is a snap-fit, the AI ​​model 120 will determine whether this clearance fit conforms to the design specifications, such as whether it meets the specified clearance requirements. Here, the design specifications can be the "Automotive General Layout Design Specifications." It is understood that the design specifications can be changed depending on the manufactured parts; for example, the design specifications can be the "Railway Freight Car Equipment Design Specifications," "Metro Car Design Specifications," etc.

[0046] In this way, the manufacturing part fitting inspection system 100 according to this embodiment first determines the connection type between multiple parts to be inspected through the artificial intelligence model 120, and judges whether the fitting relationship meets the requirements based on the connection type and design specifications. This can avoid misjudging a reasonable fitting relationship as a design problem, improve the accuracy of inspection and judgment, and the use of the artificial intelligence model 120 eliminates the need for manual inspection of the fitting relationship of each part to be inspected, thereby improving the accuracy and efficiency of the inspection of the fitting relationship of the manufacturing part and improving the efficiency of design and manufacturing.

[0047] likeFigure 1 As shown, the detection system 100 may also include a training module 130, and the artificial intelligence model 120 may be a trained artificial intelligence model. The training module 130 trains the artificial intelligence model 120 using training data to obtain a trained artificial intelligence model. In some embodiments, the training data includes text data and / or three-dimensional data. Thus, using text data can provide semantic design intent, providing a basis for decision-making for the artificial intelligence model 120; using three-dimensional data can directly display the three-dimensional structure, providing a visual basis for judgment for the artificial intelligence model 120; and using multimodal training data including text data and / or three-dimensional data to train the artificial intelligence model 120 can enable the artificial intelligence model 120 to better cope with input changes, thereby improving the robustness and stability of the artificial intelligence model 120. For example, the artificial intelligence model 120 can simultaneously use text data and three-dimensional data input to determine the connection type between multiple parts to be detected. Exemplarily, text data may include part name, part location, part shape, part connection type, design specifications of the manufactured part, etc., and three-dimensional data may include three-dimensional models, part assembly relationships, part mechanical or simulation data, etc. Here, the assembly relationship of parts refers to, for example, the positioning and alignment of parts. It is understood that the training data is not limited to this and can be changed as needed.

[0048] In some implementations, the artificial intelligence model 120 can be a generative artificial intelligence model. Generative artificial intelligence models are typically based on deep learning techniques, particularly neural network models. They learn the inherent patterns and characteristics of data through training data. Generative artificial intelligence models can not only process input data but also learn and simulate the inherent patterns of data, autonomously creating new content. For example, during training, the training data might be related to the rear bumper of a vehicle. The generative artificial intelligence model can infer and generate data related to the front bumper based on this training data, without requiring additional front bumper training data. Thus, using a generative artificial intelligence model eliminates the need to collect and label training data separately for each part in the manufactured component, reducing the amount of training data and lowering the complexity of the artificial intelligence model.

[0049] Advantageously, the generative AI model may include a semi-supervised self-learning model. The training data for the semi-supervised self-learning model includes both labeled and unlabeled sample data, reducing the model's need for large amounts of labeled sample data during the learning process and allowing it to automatically utilize unlabeled sample data to improve learning performance. In some implementations, the semi-supervised self-learning model can be trained first using labeled sample data, and then high-confidence samples from the unlabeled sample data can be automatically labeled and added to the training data to further improve model performance. Here, high confidence means that the semi-supervised self-learning model has a high confidence level in its predictions of certain unlabeled sample data; for example, if the prediction probability of certain unlabeled sample data exceeds 90%, then these unlabeled sample data are considered relatively reliable predictions. This allows the generative AI model to automatically optimize and improve its generalization ability.

[0050] In some implementations, the determination module 110 can identify multiple parts to be inspected as primary inspection parts and environmental parts, and the artificial intelligence model 120 can determine the connection type between the primary inspection parts and the environmental parts. This allows for the inspection of only the fit between the primary inspection parts and the environmental parts, reducing unnecessary inspection resource consumption and interference factors, avoiding redundant data from irrelevant parts from affecting the inspection results, and improving inspection efficiency and accuracy.

[0051] In one example, the determining module 110 can determine the main detection part based on the detection requirements, and define the parts to be detected around the main detection part as environmental parts. Specifically, the determining module 110 can define the parts to be detected within a preset distance from the main detection part as environmental parts. This preset distance can be, for example, 10mm. It is understood that this preset distance can be changed as needed and is not limited to this. For example, when the detection requirement is to detect the fit relationship of the rear bumper of a vehicle, the main detection part should be the rear bumper, and the environmental parts may include rear bumper trim, rear bumper wiring harness, rear bumper bracket, rear anti-collision beam, etc. It is understood that the main detection part and environmental parts can be changed according to the detection requirements and are not limited to this.

[0052] In another example, the determination module 110 can determine the main inspection parts and environmental parts based on a parts list pre-stored in the determination module 110. Exemplarily, the parts list may include main inspection parts and environmental parts determined based on historical inspection tasks. This allows for the assessment of the impact of design modifications on part mating relationships, for example, in the case of design changes to manufactured parts, enabling design improvements. It is understood that the content of the parts list is not limited to this; for example, the parts list may also include basic part information (such as part name, part number, etc.), part attribute information (such as material information, functional information, etc.), and part relationship information (such as assembly location, adjacent parts, etc.). Thus, for example, the main functional parts can be determined as main inspection parts based on the attribute information in the parts list, and mating parts in the parts list or parts adjacent to the main inspection parts can be determined as environmental parts, achieving rapid segmentation of main inspection parts and environmental parts and improving inspection efficiency.

[0053] Reference Figure 1 The mating detection system 100 may also include a storage module 140, which can be used to store the design specifications of the manufactured parts. In one example, the storage module 140 can be an external database or an external rule engine (e.g., Drool, Jess), and it can be integrated with the artificial intelligence model 120. After the artificial intelligence model 120 determines the connection type and detects the mating relationship, it can retrieve the design specifications of the manufactured parts by querying the external database or calling the external rule engine, thereby determining whether the detected mating relationship conforms to the design specifications of the manufactured parts. Storing the design specifications separately in the storage module 140 allows for independent maintenance and updates, improving the flexibility of the mating detection system 100. In another example, the design specifications of the manufactured parts can be stored in the artificial intelligence model 120. Specifically, during the training phase of the artificial intelligence model 120, the design specifications can be added to the training data, allowing the design specifications to be embedded in the artificial intelligence model 120 and highly coupled with it.

[0054] Advantageously, based on the complexity of the design specifications of the manufactured parts, a portion of the design specifications can be stored in the storage module 140, while another portion can be added to the training data. For example, tolerance limits, clearance requirements, and positional constraints of parts can be added to the training data as simple and relatively fixed rules to be embedded in the artificial intelligence model 120, while assembly-related mating relationships, functional and performance rules, and material strength of parts can be stored in the storage module 140 as complex and easily changing rules. In this way, the artificial intelligence model 120 can directly learn simple design rules during training, and complex design rules can be managed, updated, and maintained independently of the artificial intelligence model 120, improving the flexibility and maintainability of the mating detection system 100.

[0055] In some embodiments, one or more modules of the mating detection system 100 can be integrated into 3D design software, such as CATIA or SolidWorks, to detect the mating relationships between parts of a manufactured part. Thus, during the design phase of a manufactured part, users can use the 3D design software to check whether the mating relationships between parts conform to design specifications and make timely improvements, eliminating the need for separate detection after design completion, reducing design and modification time, and avoiding repetitive work. For example, when modifying part dimensions, the AI ​​model 120 integrated into the 3D design software can detect whether interference, gaps, etc., occur in the mating relationships related to that part, helping users adjust the design of the manufactured part in a timely manner. Furthermore, users do not need to switch to other platforms or tools, reducing workflow interruptions, quickly identifying problems in the design process, and can directly access relevant data such as text and 3D data of the manufactured part from the 3D design software without re-importing data, avoiding data loss. In addition, the detection results can be directly annotated in the 3D model of the manufactured part, allowing users to intuitively analyze mating problems in the 3D design software, facilitating the development of improvement plans.

[0056] Furthermore, the determination module 110 can be integrated into the artificial intelligence model 120. In this way, when determining the part to be inspected, relevant data of the manufactured part (e.g., 3D data, text data) can be provided to the artificial intelligence model 120 via 3D design software. The artificial intelligence model 120 determines the part to be inspected within the manufactured part and automatically performs subsequent inspection and judgment processes, thereby reducing data transmission and communication time between modules or models in the inspection system 100 and improving the inspection efficiency of the inspection system 100. It is understood that the configuration of the determination module 110 is not limited to this. For example, the determination module 110 can also be integrated into 3D design software to determine the part to be inspected within the manufactured part. This allows the use of the existing determination functions in the 3D design software, reducing the complexity of the artificial intelligence model 120 and minimizing the time required for adjustments when the artificial intelligence model 120 needs to be adapted to new tasks.

[0057] exist Figure 1In the detection system 100, a notification module 150 may be included. The notification module 150 communicates with an artificial intelligence model 120. The notification module 150 receives judgment results from the artificial intelligence model 120. If the artificial intelligence model 120 determines that the fit between multiple parts to be inspected does not conform to design specifications, the notification module 150 may send an improvement notification to the user based on the judgment result. In some embodiments, the improvement notification may include: adjusting the size or tolerance of the parts, for example, increasing the size of the rear bumper to achieve the design specifications for a clearance fit using a snap-fit ​​method; adjusting the relative position between the parts, for example, deflecting the rear bumper trim to eliminate interference with the rear bumper; changing the fit method, for example, changing the interference fit between the rear bumper and the rear bumper bracket to a transition fit to eliminate interference between the two parts. It is understood that the content of the improvement notification is not limited to these and can be modified as needed. The notification module 150 can be integrated into 3D design software to send improvement notifications to users through the 3D design software. After receiving the improvement notification, users can judge whether to improve the mating relationship between parts based on their own experience. If the user judges that improvement is needed, they can, for example, select to execute the content of the improvement notification through the 3D design software, or change the mating relationship between parts themselves based on experience. If the user judges that no improvement is needed, they can, for example, select to ignore the content of the improvement notification through the 3D design software to avoid unnecessary modifications to the design of the manufactured parts.

[0058] In some implementations, the artificial intelligence model 120 can be deployed in the cloud. Users can access the artificial intelligence model 120 via the internet. Deployed in the cloud, the artificial intelligence model 120 can utilize the powerful computing capabilities and storage resources of the cloud server to handle complex and large-scale tasks without consuming significant computing resources on the terminal, thus reducing the computational demands on the terminal. It is known that cloud servers operate in cloud computing centers. In this implementation, the cloud computing center can be established in an economically viable and secure region. Specifically, the cloud computing center can be established in a low-cost region to reduce construction and operating costs, and should be located in an area with sufficient power supply and fast, reliable communication to ensure stable operation and efficient management. Furthermore, the region where the cloud computing center is located should comply with relevant data protection regulations (e.g., the "Data Security Law of the People's Republic of China") and possess high-level security measures, including monitoring systems, access control, and firewalls, to ensure the security of data storage, transmission, and processing. In this way, the artificial intelligence model 120 can be used nationwide and even internationally through the cloud computing center, achieving resource sharing across technology centers. It is understood that the deployment method of the artificial intelligence model 120 is not limited to this; for example, the artificial intelligence model 120 can also be deployed at the edge, on terminals, etc. In some implementations, the artificial intelligence model 120 can be deployed in a hybrid manner, such as on a terminal and in the cloud, so that the terminal and the cloud can work together to improve resource utilization, without limitation.

[0059] Similarly, the notification module 150 and / or the determination module 110 can also be deployed in the cloud and / or at the edge and / or on the terminal to communicate with the artificial intelligence model 120, which will not be described in detail here.

[0060] In some implementations, the determining module 110 can partition the parts to be inspected based on the boundaries between the determined main inspection parts and the environmental parts, so as to perform partitioned inspection of the parts to be inspected. Here, the boundary is the edge limit of the part in the structure. The artificial intelligence model 120 accordingly determines the connection type between the main inspection parts and the environmental parts in each part's region, and detects the mating relationship between the main inspection parts and the environmental parts in each part's region, and determines whether the mating relationship between the main inspection parts and the environmental parts in each part's region conforms to the design specifications of the manufactured part based on the connection type. In this way, the mating relationship of the parts to be inspected can be detected in partitions, reducing interference from irrelevant areas, more accurately analyzing the mating situation of parts in each part's region, clearly displaying the mating status of each region, making the mating problems in key areas more obvious, and facilitating design adjustments.

[0061] Combination Figure 2 and Figure 3As shown, in an exemplary embodiment of the present invention, the determining module 110 determines that the main detection part among the parts to be detected is the rear bumper 210 of the vehicle, and the environmental parts are the rear bumper trim 220, the rear bumper wiring harness 230, the rear bumper bracket 240, and the rear anti-collision beam 250. The determining module 110 divides the parts to be detected into four regions according to the boundary between the main detection part and the environmental parts, namely the first region A, the second region B, the third region C, and the fourth region D. Artificial intelligence model 120 determines the connection type of the main detection parts and environmental parts in different areas, as shown in Table 1 below. In the first area A, the rear bumper 210 and the rear bumper trim 220 are connected by plastic snap-fit ​​type 2 (e.g., structural snap-fit). In the second area B, the rear bumper 210 and the rear bumper wiring harness 230 are connected by plastic snap-fit ​​type 1 (e.g., wiring harness fixing snap-fit). In the third area C, the rear bumper 210 and the rear bumper bracket 240 are welded. In the fourth area D, there is no connection between the rear bumper 210 and the rear anti-collision beam 250. Taking the first region A as an example, according to the design specifications, the gap requirement for the snap-fit ​​position of plastic snap-fit ​​type 2 is 0mm. The artificial intelligence model 120 detected that the gap of the snap-fit ​​position in the first region A is -1mm, which does not meet the design specifications. The artificial intelligence model 120 can output the judgment result "NOK", indicating that the fit relationship of the snap-fit ​​position in the first region A does not meet the design specifications. In addition, the gap requirement for the non-snap-fit ​​position is greater than 1mm. The artificial intelligence model 120 detected that the gap of the non-snap-fit ​​position in the first region A is 2mm, which meets the design specifications. The artificial intelligence model 120 can output the judgment result "OK", indicating that the fit relationship of the non-snap-fit ​​position in the first region A meets the design specifications.

[0062] Table 1. Rear Bumper Fit Test Table

[0063]

[0064]

[0065] Based on the concept of this invention, a method for testing the fit of manufactured parts is also provided. (Refer to...) Figure 4 A method for testing the fit of manufactured parts according to an embodiment of the present invention includes the following steps:

[0066] Step S101: Identify multiple parts to be inspected in the manufactured parts based on inspection requirements.

[0067] Step S102: Based on the identified multiple parts to be inspected, determine the connection type between the parts to be inspected using an artificial intelligence model.

[0068] Step S103: Detect the mating relationship between multiple parts to be tested using an artificial intelligence model.

[0069] Step S104: Based on the connection type, use an artificial intelligence model to determine whether the mating relationship conforms to the design specifications.

[0070] Step S105: If the judgment result is "Y", that is, if the mating relationship meets the design specifications, the corresponding judgment result is output, and the user can continue the design process of manufacturing the part based on the judgment result.

[0071] Step S106: If the judgment result is "N", that is, if the fit relationship does not meet the design specifications, the corresponding judgment result is output. The user can improve the fit relationship of the parts in the manufactured part based on the judgment result.

[0072] Please refer to the above text for the function of the manufacturing part fit inspection method according to this embodiment. Figure 1 The content mentioned above will not be repeated here.

[0073] Reference Figure 5 According to another embodiment of the present invention, a method for detecting the fit of a manufactured part includes the following steps:

[0074] Step S201: Train the artificial intelligence model using training data.

[0075] Step S202: Based on the inspection requirements, identify multiple parts to be inspected in the manufactured parts, and determine the parts to be inspected as the main inspection parts and environmental parts.

[0076] Step S203: Divide the parts to be inspected into zones according to the determined boundaries between the main inspection parts and the environmental parts.

[0077] Step S204: Determine the connection type between the main detection part and the environmental parts in the area where each part is located using an artificial intelligence model.

[0078] Step S205: Detect the mating relationship between the main detection part and the environmental parts in the area where each part is located using an artificial intelligence model.

[0079] Step S206: Based on the connection type, use an artificial intelligence model to determine whether the mating relationship between the main detection part and the environmental parts in the area where each part is located conforms to the design specifications of the manufacturing part.

[0080] Step S207: If the judgment result is "Y", that is, the fit relationship between the parts to be inspected meets the design specifications, output the judgment result and end the inspection process.

[0081] Step S208: If the judgment result is "N", that is, the fit relationship between the parts to be tested does not meet the design specifications, an improvement notice is sent to the user to improve the fit relationship.

[0082] Please refer to the above text for the function of the manufacturing part fit inspection method according to this embodiment. Figure 1 The content mentioned above will not be repeated here.

[0083] Reference Figure 6 According to another embodiment of the present invention, a method for detecting the fit of a manufactured part includes the following steps:

[0084] Step S301: Train the artificial intelligence model using training data.

[0085] Step S302: Based on the inspection requirements, determine the main inspection parts in the manufactured parts, and determine the parts within a preset distance from the main inspection parts as environmental parts.

[0086] Step S303: Determine the connection type between the main detection part and the environmental parts using an artificial intelligence model.

[0087] Step S304: Detect the mating relationship between the main detection part and the environmental parts using an artificial intelligence model.

[0088] Step S305: Based on the connection type, use an artificial intelligence model to determine whether the mating relationship conforms to the design specifications of the manufactured part.

[0089] Step S306: If the judgment result is "Y", that is, the fit relationship between the parts to be inspected meets the design specifications, output the inspection result and end the inspection process.

[0090] Step S307: If the judgment result is "N", that is, the fit relationship between the parts to be tested does not meet the design specifications, send an improvement notice to the user to improve the fit relationship.

[0091] Please refer to the above text for the function of the manufacturing part fit inspection method according to this embodiment. Figure 1 The content mentioned above will not be repeated here.

[0092] Reference Figure 7 According to another embodiment of the present invention, a method for detecting the fit of a manufactured part includes the following steps:

[0093] Step S401: Train the artificial intelligence model using training data.

[0094] Step S402: Determine the main inspection parts and environmental parts based on the pre-stored parts list.

[0095] Step S403: Determine the connection type between the main detection part and the environmental parts using an artificial intelligence model.

[0096] Step S404: Detect the mating relationship between the main detection part and the environmental parts using an artificial intelligence model.

[0097] Step S405: Based on the connection type, use an artificial intelligence model to determine whether the mating relationship conforms to the design specifications of the manufactured part.

[0098] Step S406: If the judgment result is "Y", that is, the fit relationship between the parts to be inspected meets the design specifications, output the inspection result and end the inspection process.

[0099] Step S407: If the judgment result is "N", that is, the fit relationship between the parts to be tested does not meet the design specifications, send an improvement notice to the user to improve the fit relationship.

[0100] Please refer to the above text for the function of the manufacturing part fit inspection method according to this embodiment. Figure 1 The content mentioned above will not be repeated here.

[0101] The present invention also provides a computer device, with reference to Figure 8 , Figure 8 This is a schematic diagram of a computer device 300 according to one embodiment of the present invention.

[0102] like Figure 8 As shown, the computer device 300 may include a memory 310 and a processor 320. The memory 310 may store instructions 311, which may be executed by the processor 320. When the processor 320 executes the instructions 311, it implements the manufacturing part fitting detection method according to the above embodiment.

[0103] The computer device 300 in this embodiment can be a laptop, desktop computer, or cloud server, etc. It is understood that the components included in the computer device 300 are not limited to the memory 310 and processor 320, and can vary depending on different needs. Exemplarily, the computer device 300 may also include multiple components connected to its input / output interfaces (…). Figure 8 (Not shown in the image), including but not limited to: input units, such as keyboards, mice, etc.; output units, such as displays, speakers, etc.; storage units, such as semiconductor storage devices, magnetic surface storage devices, optical storage devices, etc.; and communication modules, such as network interface cards, wireless communication transceivers, etc.

[0104] In some implementations, memory 310 may include, for example, random access memory (RAM) or read-only memory (ROM). Memory 310 may be used to store instructions, programs, code, and other programs and data required by the computer device, but is not limited thereto.

[0105] In addition, the processor 320 can be a central processing unit (CPU) or other general-purpose processors, such as digital signal processing (DSP), field-programmable gate array (FPGA), programmable logic array (PLA), etc.

[0106] In an exemplary embodiment of the present invention, a computer-readable storage medium is also provided having computer-executable instructions stored thereon for performing a mating detection method for a manufactured part according to the above embodiments.

[0107] Alternatively, the computer-readable storage medium according to this embodiment may be a ROM, RAM, semiconductor storage device, magnetic surface storage device, and optical storage device, etc.

[0108] The present invention also proposes a computer program product comprising computer-executable instructions that, when executed, cause at least one processor to perform the mating detection method for a manufactured part according to the above embodiments.

[0109] Generally, various embodiments of the present invention can be implemented in hardware, dedicated circuitry, software programs, firmware, logic circuitry, or any combination thereof, as needed. Specifically, some aspects may be implemented in hardware, while others may be implemented in firmware or software programs executable by a controller, microprocessor, or other computing device. When aspects of embodiments of the present invention are illustrated or described as block diagrams, flowcharts, or using some other graphical representation, it will be understood that the blocks, apparatuses, systems, techniques, or methods described herein can be implemented as non-limiting examples in hardware, software, firmware, dedicated circuitry, logic circuitry, general-purpose hardware, or controllers or other computing devices, or some combination thereof.

[0110] Computer-readable program instructions or computer program products for executing various embodiments of the present invention can also be stored in the cloud. When needed, users can access the computer-readable program instructions stored in the cloud for executing an embodiment of the present invention via mobile internet, fixed network or other networks, thereby implementing various embodiments of the present invention.

[0111] It should be understood that the embodiments shown in the figures only illustrate optional configurations of the mating detection system for manufactured parts according to the present invention; however, they are merely illustrative and not limiting. Other configurations may be adopted without departing from the spirit and scope of the present invention.

[0112] The technical content and features of the present invention have been disclosed above. However, it is understood that those skilled in the art can make various changes and improvements to the disclosed concepts under the inventive concept of the present invention, all of which fall within the protection scope of the present invention. The description of the above embodiments is illustrative rather than restrictive, and the protection scope of the present invention is determined by the claims.

Claims

1. A mating inspection system for manufactured parts, characterized in that, The matching detection system (100) includes: A determination module (110) is configured to determine a plurality of parts to be inspected in the manufactured part based on inspection requirements; An artificial intelligence model (120) is configured to determine the connection type between the plurality of parts to be detected based on the determined plurality of parts to be detected. The artificial intelligence model (120) is further configured to: detect the mating relationship between the plurality of parts to be detected, and determine whether the mating relationship conforms to the design specifications of the manufactured part based on the connection type.

2. The matching detection system according to claim 1, characterized in that, The artificial intelligence model (120) is a trained artificial intelligence model, and the coordination detection system (100) includes a training module (130), which is configured to train the artificial intelligence model (120) with training data, wherein the training data includes text data and / or three-dimensional data.

3. The matching detection system according to claim 2, characterized in that, The text data includes at least a portion of the part name and / or the area where the part is located and / or the part shape and / or the part connection type and / or the design specifications of the manufactured part, and the three-dimensional data includes the three-dimensional model of the part and / or the assembly relationship of the part and / or the mechanical or simulation data of the part.

4. The matching detection system according to claim 1, characterized in that, The determining module (110) is configured to determine the plurality of parts to be tested as the main testing parts and the environmental parts.

5. The matching detection system according to claim 4, characterized in that, The determining module (110) is configured to determine the main detection part based on the detection requirements, and to determine the part to be detected within a preset distance from the main detection part as the environmental part.

6. The mating detection system according to claim 4, characterized in that, The determining module (110) is configured to determine the main detection part and the environmental part based on a parts list, wherein the parts list is pre-stored in the determining module (110).

7. The mating detection system according to claim 4, characterized in that, The determining module (110) is configured to partition the plurality of parts to be detected according to the determined boundaries of the main detection part and the environmental part; The artificial intelligence model (120) is configured to determine the connection type between the main detection part and the environmental part in the area where each part is located, and to detect the mating relationship between the main detection part and the environmental part in the area where each part is located, and to determine whether the mating relationship between the main detection part and the environmental part in the area where each part is located conforms to the design specifications of the manufactured part based on the connection type.

8. The matching detection system according to claim 1, characterized in that, The mating detection system (100) includes a storage module (140) configured to store at least a portion of the design specifications of the manufactured part, and / or the storage module (140) is integrated with the artificial intelligence model (120).

9. The matching detection system according to claim 1, characterized in that, The mating detection system (100) includes a notification module (150) configured to send an improvement notification to a user when the mating relationship between the plurality of parts to be tested does not conform to the design specifications of the manufactured part, so that the user can improve the mating relationship.

10. The matching detection system according to claim 9, characterized in that, The notification module (150) is deployed in the cloud and / or at the edge and / or on the terminal, and is integrated into the 3D design software to send improvement notifications to the user through the 3D design software.

11. The mating detection system according to any one of claims 1 to 10, characterized in that, The determining module (110) is integrated into the artificial intelligence model (120), or the determining module (110) is integrated into the three-dimensional design software.

12. The mating detection system according to any one of claims 1 to 10, characterized in that, The artificial intelligence model (120) and / or the determining module (110) are deployed in the cloud and / or at the edge and / or on the terminal.

13. The mating detection system according to any one of claims 1 to 10, characterized in that, The artificial intelligence model (120) is a generative artificial intelligence model, which includes a semi-supervised self-learning model.

14. The mating detection system according to any one of claims 1 to 10, characterized in that, The artificial intelligence model (120) is integrated into the 3D design software for cooperative testing.

15. The mating detection system according to any one of claims 1 to 10, characterized in that, The fitting relationship includes clearance fit and / or interference fit, and / or the connection type includes snap-fit ​​and / or welding and / or threaded connection and / or riveting.

16. A method for testing the fit of manufactured parts, characterized in that, The matching detection method includes: Based on the testing requirements, a plurality of parts to be tested in the manufactured parts are determined (S101); Based on the identified multiple parts to be tested, the connection type between the multiple parts to be tested is determined by an artificial intelligence model (S102); The mating relationship between the multiple parts to be tested is detected by an artificial intelligence model (S103); Based on the connection type, an artificial intelligence model is used to determine whether the mating relationship conforms to the design specifications of the manufactured part (S104; S305; S405).

17. The method for detecting the fit according to claim 16, characterized in that, The matching detection method includes: training the artificial intelligence model using training data (S201; S301; S401).

18. The method for detecting the fit according to claim 16, characterized in that, Determining a plurality of parts to be inspected in the manufactured part includes: identifying the plurality of parts to be inspected as main inspection parts and environmental parts (S202).

19. The method for detecting the fit according to claim 18, characterized in that, Determining multiple parts to be inspected in the manufactured part includes: determining the main inspection part based on inspection requirements, and determining the parts to be inspected within a preset distance from the main inspection part as the environmental parts (S302).

20. The method for detecting the fit according to claim 18, characterized in that, Determining multiple parts to be inspected in the manufactured part includes: determining the main inspection part and the environmental part based on a pre-stored parts list (S402).

21. The method for detecting the fit according to claim 18, characterized in that, The matching detection method includes: Based on the determined boundaries between the main detection component and the environmental component, the plurality of components to be detected are partitioned (S203); The connection type between the main detection part and the environmental part in the area where each part is located is determined by the artificial intelligence model (S204); The artificial intelligence model is used to detect the cooperation relationship between the main detection part and the environmental part in the area where each part is located (S205); Based on the connection type, the artificial intelligence model determines whether the fit between the main detection part and the environmental part in the area where each part is located conforms to the design specifications of the manufactured part (S206).

22. The method for detecting fit according to any one of claims 16 to 21, characterized in that, The mating detection method includes: sending an improvement notification to the user when the mating relationship between the plurality of parts to be tested does not conform to the design specifications of the manufactured part, so that the user can improve the mating relationship (S208; S307; S407).

23. A computer device, characterized in that, The computer device (300) includes a memory (310), a processor (320), and instructions (311) stored in the memory (310) and executable by the processor (320), wherein the processor (320) executes the instructions (311) to implement the mating detection method for manufactured parts according to any one of claims 16 to 22.

24. A computer-readable storage medium, characterized in that, The computer-readable storage medium has computer-executable instructions stored thereon for performing the mating detection method for a manufactured part according to any one of claims 16 to 22.

25. A computer program product comprising computer-executable instructions, characterized in that, When the computer-executable instructions are executed by at least one processor, the method for detecting the fit of a manufactured part according to any one of claims 16 to 22 is implemented.

Images