Car body connection process optimization method and system based on multi-physics field model

By optimizing the vehicle body connection process using a multiphysics model, and combining virtual digital twin models and image processing technology, weak points are identified and optimized, which solves the limitations and reliance on experience in traditional methods and improves the quality and stability of vehicle body connections.

CN118395596BActive Publication Date: 2024-11-19GESTAMP AUTO COMPONENTS BEIJING CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410572003.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-05-10
Publication Date
2024-11-19
Estimated Expiration
2044-05-10

AI Technical Summary

Technical Problem

Traditional methods for optimizing vehicle body connection processes are limited to specific scenarios or parameters, lack a global optimization perspective, and rely on experience and trial and error, resulting in poor stability and reliability of the results.

Method used

A multi-physics model-based method for optimizing the vehicle body connection process is adopted. By establishing a virtual digital twin model and combining image acquisition and adaptive enhancement processing technology, an application scenario dataset is obtained. Multiple physics models are configured for stress fitting, weak points in the connection are identified, and production parameters are optimized.

Benefits of technology

It enables comprehensive modeling and analysis of the vehicle body connection process, improves the scientific nature and accuracy of the optimization process, and can specifically improve weak links, thereby enhancing connection quality and stability.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118395596B_ABST
    Figure CN118395596B_ABST
Patent Text Reader

Abstract

The present invention provides a method and system for optimizing the vehicle body connection process based on a multi-physics field model, which relates to the field of image data processing technology, including: establishing a proofing sample; establishing a virtual digital twin model; performing image acquisition, establishing an image set of the proofing sample; inputting into an image processing network, performing adaptive enhancement processing of the image, and extracting the connection features of the proofing sample; synchronizing to the virtual digital twin model according to the position identifier, and updating the virtual digital twin model; obtaining an application scenario data set, configuring M physical field models, performing force fitting of the updated virtual digital twin model, and establishing a force fitting result; constructing connection weak points, and performing parameter optimization of production parameters. The present invention solves the technical problems that traditional methods are often limited to the optimization of a specific scenario or specific parameters, lack a global optimization perspective, and usually rely on experience and trial and error, resulting in poor stability and reliability of the final result.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of image data processing, and in particular to a method and system for optimizing a vehicle body connection process based on a multi-physics field model. Background Art

[0002] The body connection process directly affects important performance indicators such as the structural strength, stiffness and durability of the vehicle. Optimizing the body connection process can improve the overall quality and stability of the body structure and improve the safety of the vehicle. However, there are still some problems with the traditional body connection process optimization method. On the one hand, traditional methods are often limited to the optimization of a specific scenario or specific parameters, and lack a global optimization perspective, which leads to the optimization results being effective only in certain specific cases, but not in other cases; on the other hand, traditional optimization methods usually rely on experience and trial and error, lack scientific theoretical support and systematic optimization processes, which makes the optimization process time-consuming and labor-intensive, and cannot guarantee the stability and reliability of the final results. Summary of the invention

[0003] This application provides a body connection process optimization method based on a multi-physics field model, aiming to solve the technical problems that traditional body connection process optimization methods are often limited to the optimization of a specific scenario or specific parameters, lack a global optimization perspective, and usually rely on experience and trial and error, resulting in poor stability and reliability of the final result.

[0004] In view of the above problems, the present application provides a method and system for optimizing the car body connection process based on a multi-physics field model.

[0005] The first aspect disclosed in the present application provides a method for optimizing a vehicle body connection process based on a multi-physical field model, the method comprising: establishing a proofing sample, wherein the proofing sample is a connection sample of a vehicle body, and the proofing sample is provided with identifications of production parameters and sample data; establishing a virtual digital twin model with the production parameters and the sample data, wherein the virtual digital twin model is an established mathematical analysis model; performing image acquisition on the proofing sample to establish an image set of the proofing sample, wherein the image set has a position identification, and the image set includes a digital image set and a radiographic image set; inputting the image set into an image processing network, performing adaptive enhancement processing of the image, and extracting connection features of the proofing sample; synchronizing the connection features to the virtual digital twin model according to the position identification, thereby updating the virtual digital twin model; obtaining an application scenario data set, configuring M physical field models based on the application scenario data set, performing force fitting of the updated virtual digital twin model with the M physical field models, and establishing a force fitting result, wherein M is an integer greater than 1; constructing connection weak points through the force fitting results, and performing parameter optimization of production parameters through the connection weak points.

[0006] The second aspect disclosed in the present application provides a vehicle body connection process optimization system based on a multi-physics field model, and the system is used for the above-mentioned vehicle body connection process optimization method based on a multi-physics field model, and the system includes: a sample establishment module, the sample establishment module is used to establish a proofing sample, wherein the proofing sample is a connection sample of the vehicle body, and the proofing sample is provided with an identifier of production parameters and sample data; a model establishment module, the model establishment module is used to establish a virtual digital twin model with the production parameters and the sample data, wherein the virtual digital twin model is an established mathematical analysis model; an image acquisition module, the image acquisition module is used to perform image acquisition on the proofing sample, and establish an image set of the proofing sample, wherein the image set has a position identifier, and the image set includes a digital image set, a shot A line image set; an image enhancement processing module, the image enhancement processing module is used to input the image set into an image processing network, perform adaptive enhancement processing of the image, and extract connection features of the proofing sample; a model update module, the model update module is used to synchronize the connection features to the virtual digital twin model according to the position identifier, so as to update the virtual digital twin model; a force fitting module, the force fitting module is used to obtain an application scenario data set, configure M physical field models based on the application scenario data set, perform force fitting of the updated virtual digital twin model with the M physical field models, and establish a force fitting result, wherein M is an integer greater than 1; a parameter optimization module, the parameter optimization module is used to construct connection weak points through the force fitting results, and perform parameter optimization of production parameters through the connection weak points.

[0007] According to a third aspect disclosed in the present application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, any step of the first aspect disclosed in the present application is implemented.

[0008] The fourth aspect disclosed in the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the first aspect disclosed in the present application.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] By establishing a virtual digital twin model, combining the sample data and production parameters in actual production with the mathematical analysis model, the comprehensive modeling and analysis of the body connection process is realized, and the scientific nature of the optimization process is improved; the image acquisition and adaptive enhancement processing technology can accurately capture the key features of the body connection process, which provides real-time data support and provides more accurate and comprehensive information for the optimization process; by obtaining the application scenario data set and configuring multiple physical field models based on the data set, the multi-faceted optimization of the virtual digital twin model is realized, which makes the optimization process more comprehensive and can fully consider the influence of different physical scenarios on the connection process; the connection weak points are constructed through the force fitting results, and these weak points are used to optimize the parameters of the production parameters, which enables the optimization process to improve the weak links of the connection process in a targeted manner and improve its quality and stability. In summary, this method effectively solves the limitations of the traditional body connection process optimization, improves the scientific nature, accuracy and comprehensiveness of the optimization process, and thus achieves the technical effect of improving the quality and stability of the body connection process.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic flow chart of a method for optimizing a vehicle body connection process based on a multi-physics field model provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of the structure of a vehicle body connection process optimization system based on a multi-physics field model provided in an embodiment of the present application;

[0014] Figure 3 This is a diagram of the internal structure of a computer device provided in an embodiment of the present application.

[0015] Explanation of the reference numerals: sample building module 10 , model building module 20 , image acquisition module 30 , image enhancement processing module 40 , model updating module 50 , force fitting module 60 , parameter optimization module 70 . DETAILED DESCRIPTION

[0016] The embodiments of the present application provide a vehicle body connection process optimization method based on a multi-physics field model, thereby solving the technical problems that traditional vehicle body connection process optimization methods are often limited to the optimization of a specific scenario or specific parameters, lack a global optimization perspective, and usually rely on experience and trial and error, resulting in poor stability and reliability of the final results.

[0017] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0018] like Figure 1 As shown, the embodiment of the present application provides a method for optimizing a vehicle body connection process based on a multi-physics field model, the method comprising:

[0019] Establishing a proofing sample, wherein the proofing sample is a connection sample of the vehicle body, and the proofing sample is provided with identifications of production parameters and sample data;

[0020] Establishing proofing samples includes collecting relevant information about body connections, including connections of different materials, connection technology, parameter optimization, cost efficiency deformation, etc. This information is used to build a virtual model to simulate the actual body connection. Specifically, first collect various sample data of body connections, including but not limited to connection technology, connection materials, connection parts, production parameters, etc. These data come from actual production processes, historical data, laboratory tests, etc. Then, determine the production parameters related to body connections, such as welding temperature, pressure, welding time, riveting pressure, etc. These parameters will affect the quality and performance of the connection.

[0021] Each proof sample is identified in order to associate it with specific production parameters and sample data. This can be a digital ID or other form of identifier to facilitate identification in subsequent steps.

[0022] Establishing a virtual digital twin model with the production parameters and the sample data, wherein the virtual digital twin model is an established mathematical analysis model;

[0023] Based on the collected production parameters and sample data, a virtual digital twin model is established using mathematical methods. This mathematical model is a computer simulation-based model that simulates the behavior and performance of the body connection. Specifically, CAD software or other simulation modeling tools are used to parameterize the production parameters and sample data with the mathematical model, so that the model can generate corresponding outputs based on different parameter inputs, thereby simulating the body connection under different conditions.

[0024] Capturing images of the proofing sample to establish an image set of the proofing sample, wherein the image set has a position identifier and includes a digital image set and a radiographic image set;

[0025] Prepare equipment for image acquisition, such as high-resolution cameras, X-ray machines, etc., place proofing samples in the image acquisition area, shoot or scan the proofing samples to obtain their appearance features, obtain digital image sets through high-resolution cameras, obtain radiographic image sets through X-ray machines, integrate the image sets, and establish an image set of the proofing samples, in which each image is associated with the corresponding sample position and marked.

[0026] Inputting the image set into an image processing network, performing adaptive enhancement processing on the image, and extracting connection features of the proofing samples;

[0027] Construct an image processing network, whose main function is to extract connection features from the preprocessed image. This network can be a convolutional neural network (CNN) based on deep learning, a feature extractor, and other models. The preprocessed image set is used as input to the constructed image processing network. In the image processing network, adaptive enhancement processing is performed. This processing method can automatically adjust parameters according to the content and features of the image to maximize the quality and information richness of the image. Connection features are extracted from the image after adaptive enhancement processing. These features cover information such as the shape, structure, and material properties of the connection parts, and the feature extraction results are output, including the extracted connection features and the corresponding sample position identifiers.

[0028] Synchronizing the connection features to the virtual digital twin model according to the location identifier, thereby updating the virtual digital twin model;

[0029] The extracted connection features are synchronized to the virtual digital twin model according to their location identifiers, so as to associate the connection features with the corresponding positions in the virtual digital twin model, update the features to the model, and update the virtual digital twin model based on the synchronously updated connection features so that the model can more accurately reflect the actual situation.

[0030] Acquire an application scenario data set, configure M physical field models based on the application scenario data set, perform force fitting of the updated virtual digital twin model with the M physical field models, and establish a force fitting result, where M is an integer greater than 1;

[0031] Obtain application scenario data sets, which contain actual application scenario data of various vehicle body connections, such as driving conditions under various road conditions, temperature and humidity in different environments, and other information. The purpose of obtaining these data sets is to simulate the forces exerted on vehicle body connections under different conditions.

[0032] Based on the application scenario data set, M physical field models are configured, where M is the number of physical field models and can be set according to actual conditions and the number of data sets. These physical field models are set according to factors such as different stress sources and different environmental conditions to simulate different stress conditions. The purpose of configuring these physical field models is to simulate and analyze the stress conditions of the vehicle body connection under different conditions.

[0033] Using these M physical field models, the updated virtual digital twin model is subjected to force fitting, which means that the simulated force conditions are applied to the virtual digital twin model to simulate the force conditions of the vehicle body connection in different scenarios. This can be achieved by taking the output of the physical field model as input, combining the structure and parameters of the virtual digital twin model, and simulating the force fitting to finally establish the force fitting results. These results reflect the force conditions of the vehicle body connection in different application scenarios, including information such as the force distribution and stress distribution of each connection part. These results are used to evaluate the strength and stability of the connection structure, and provide a basis for optimizing the vehicle body connection process.

[0034] The connection weak points are constructed through the force fitting results, and the parameters of the production parameters are optimized through the connection weak points.

[0035] Weak points in connection refer to parts that are prone to failure or performance degradation under stress, which may be caused by factors such as material selection, connection process, and design structure. The force fitting results obtained are analyzed to identify the specific locations and characteristics of weak points in the body connection. These weak points can be manifested as concentrated force, excessive stress, severe deformation, etc. For the identified weak points in connection, the relevant production parameters are optimized, including the adjustment of welding parameters, optimization of material selection, and improvement of connection process. The goal of optimization is to improve the performance of weak points in connection and improve the quality and reliability of the overall connection.

[0036] Furthermore, the step of inputting the image set into an image processing network, performing adaptive enhancement processing on the image, and extracting connection features of the proofing samples includes:

[0037] Performing image classification of the image set by a preprocessing unit to establish a digital image set;

[0038] Performing distortion correction on the digital image set based on acquisition parameters, and arranging and splicing the distortion correction results according to the position identifiers to generate an arrangement and splicing result;

[0039] The arrangement and splicing result is subjected to perspective recognition, the perspective recognition result is subjected to enhancement processing based on an image enhancement unit, and the enhancement processing result is output.

[0040] The mixed images in the image set are classified by the preprocessing unit, and the digital images are separated to establish a digital image set. The acquisition parameters used in image acquisition, including the camera's internal and external parameters, are obtained. According to the acquisition parameters, the digital image set is subjected to distortion correction. The distortion correction methods include lens distortion correction, perspective distortion correction, etc., which aim to eliminate the distortion effect in the image and make the image more realistic and accurate.

[0041] Determine the position mark of each corrected image on the vehicle body, where at least two images are taken at each position, arrange and splice the distortion-corrected images according to their position marks, and generate an arrangement and splicing result. This result is a complete image set, and each position point contains multiple distortion-corrected images and corresponding position mark information.

[0042] Perspective recognition is performed on the arrangement and stitching results. Perspective recognition is an image processing technology that aims to identify perspective deformation or distortion in images. It can detect perspective deformation or distortion by identifying geometric shapes in the image, such as straight lines and planes, and infer the perspective effect by analyzing the position and relationship of these geometric shapes in the image. Among them, perspective deformation is caused by the relative position relationship between the camera and the photographed object, which causes the object to present a perspective effect in the image, that is, the visual experience of being larger when near and smaller when far away; perspective distortion is caused by the characteristics of the optical system, such as lens distortion, perspective distortion, etc., which will cause the straight lines or planes in the image to bend or distort.

[0043] Enhancement processing based on the image enhancement unit is a technology used to improve image quality, enhance image features and details. Image enhancement technologies include contrast enhancement, sharpening, denoising, color correction, etc. Applying the image enhancement unit on the perspective recognition result can further optimize the image, reduce the impact of perspective deformation or perspective distortion on image quality, and make the image clearer and more realistic. Based on the image enhancement unit, the image is enhanced to obtain the best image enhancement effect. Through repeated trials and parameter adjustments, the most suitable parameter combination is found to achieve the best quality of the enhanced image. The enhanced image result is output. This output result reflects the final effect of the perspective recognition result after enhancement processing.

[0044] Furthermore, the step of performing distortion correction on the digital image set based on acquisition parameters, and arranging and splicing the distortion correction results according to the position identifiers to generate the arrangement and splicing results includes:

[0045] Perform initial arrangement and splicing based on the position identifier to establish a basic position splicing result;

[0046] Performing feature alignment search at the same position through the basic position splicing result, wherein the feature alignment search is a process of performing feature search and feature alignment within a preset tolerant anchor frame;

[0047] Based on the feature alignment search, the basic position stitching result is corrected and compensated to generate the arrangement stitching result.

[0048] Using the determined position markers, the images are initially arranged and stitched so that the images can be correctly aligned and stitched during the arrangement and stitching process. After the initial arrangement and stitching is completed, the basic position stitching result is established. This result reflects the initial arrangement of the images during the arrangement and stitching process, that is, the preliminary arrangement and alignment of the images at each position point.

[0049] Using the established basic position stitching result as a starting point, for each position in the basic position stitching result, a feature alignment search is performed in different images at the position. Specifically, when performing the feature alignment search, a preset tolerant anchor frame is set to ensure that the search range is sufficient to cover the features that may exist at the same position. Feature detection algorithms such as SIFT and SURF are used to search for corresponding feature points in the preset tolerant anchor frame. For the feature points detected at each position, feature alignment is performed. The feature alignment search includes steps such as feature point matching and feature alignment to ensure that the corresponding features in different images at the same position can be correctly aligned.

[0050] According to the results of the feature alignment search, the basic position stitching results are corrected, including adjusting the image position, rotation angle, scaling ratio, etc., so that the features are aligned between different images. According to the corrected basic position stitching results, the final arrangement stitching results are generated. This result reflects the final arrangement of the image after feature alignment correction during the arrangement stitching process.

[0051] Furthermore, the feature alignment search at the same position is performed through the basic position splicing result, including:

[0052] Set similarity matching threshold;

[0053] For any feature alignment search, if the similarity matching threshold of the feature satisfies the similarity matching threshold, the moving direction and moving distance of the corresponding feature alignment search are recorded as calibration data;

[0054] Arrange all calibration data, establish a calibration data set, and perform alignment correction on feature alignment search that does not meet the similarity matching threshold based on the calibration data set;

[0055] The correction compensation of the basic position stitching result is completed according to the alignment correction result.

[0056] Set a similarity matching threshold, which is used to indicate the required similarity of feature matching. The higher the threshold, the higher the required matching similarity.

[0057] For the feature points at each position, a feature alignment search is performed within the preset tolerant anchor frame. For each feature alignment search, the similarity between the feature points is calculated, which can be done by comparing the distance or other similarity metrics between them. If the calculated similarity meets the preset similarity matching threshold, the match is considered successful, and the moving direction and moving distance of the corresponding feature alignment search are recorded as calibration data.

[0058] All the recorded calibration data are sorted and summarized to form a complete calibration data set, which includes the moving direction and moving distance of each successful feature alignment search. For feature alignment searches that do not meet the similarity matching threshold, the established calibration data set is used for alignment correction. Specifically, for example, there are image 1 and image 2, image 1 is the underlying image, let it not move, adjust image 2, and adjust the feature points of image 2 accordingly according to the moving direction and moving distance in the calibration data set to align them with the corresponding feature points in the underlying image. The corrected feature points are applied to the image to correct the results of the feature alignment search that does not meet the similarity matching threshold. This can improve the quality of image stitching and make the features more consistent and accurate between different images.

[0059] According to the results of the alignment correction, the basic position stitching result is corrected and compensated. Specifically, according to the moving direction and distance of each feature point, its position in the basic position stitching result is adjusted to align it with the corresponding feature points in other images. The corrected and compensated feature points are updated to the basic position stitching result to replace the original position information, and the corrected basic position stitching result is obtained, which reflects the accurate position of the feature points after the alignment correction.

[0060] Furthermore, the alignment correction of the feature alignment search that does not meet the similarity matching threshold based on the calibration data set also includes:

[0061] Performing cluster analysis of movement direction and movement distance based on the calibration data set, and generating a first movement reference vector based on a maximum clustering result;

[0062] Performing positional proximity matching on the feature alignment search that does not meet the similarity matching threshold through the maximum clustering result, and constructing a second movement reference vector according to the positional proximity matching result;

[0063] A reference coefficient of the first moving reference vector and the second moving reference vector is configured, and alignment correction is performed based on the reference coefficient, the first moving reference vector, and the second moving reference vector.

[0064] Cluster analysis is performed based on the calibration data set, which can adopt various clustering algorithms, such as K-means clustering, hierarchical clustering, etc. The purpose of cluster analysis is to divide the moving direction and moving distance in the calibration data set into several categories to find the patterns therein.

[0065] The calibration data set is clustered. The algorithm divides the data into different clusters according to their similarities and determines the center point of each cluster. According to the clustering results, the center point of each cluster is selected as the representative movement vector of the cluster, which reflects the main trends of different movement directions and movement distances in the calibration data set. The maximum clustering result among all clustering results is obtained, and the representative movement vector of the clustering result is obtained as the first movement reference vector.

[0066] For feature alignment searches that do not meet the similarity matching threshold, based on their position information, search for feature alignment searches that are close to their positions in the maximum clustering results. Close positions refer to feature pairs with close distances. The second moving reference vector is constructed based on the matching results of close positions. Specifically, the second moving reference vector is generated based on the moving direction and moving distance of the feature alignment search of close position matching.

[0067] The reference coefficients are configured for the first moving reference vector and the second moving reference vector. The reference coefficients may be weight values, which are used to adjust the importance of the first and second moving reference vectors in the alignment correction. These coefficients may be configured according to actual conditions and requirements. Based on the configured reference coefficients, the first moving reference vector and the second moving reference vector are weightedly combined, for example, by using a weighted average method to combine the two reference vectors to obtain a final alignment correction result, thereby improving the accuracy of feature alignment.

[0068] Furthermore, the parameter optimization of production parameters through the connection weak points includes:

[0069] Based on the weak connection point, control optimization is performed under the same production process to generate a first process optimization result;

[0070] Based on the weak connection point, control optimization is performed under process improvement to generate a second process optimization result;

[0071] A cost-effective balance analysis is performed on the first process optimization result and the second process optimization result, and parameter optimization is completed based on the balance analysis results.

[0072] The main goal of control optimization under the same production process is to improve the quality and stability of the body connection by adjusting the existing production process, optimizing parameter settings or improving connection technology. Specifically, analyze the weak points of the connection, including structural problems, material selection, process flow, etc. of the connection parts to determine the existing problems and potential room for improvement. According to the analysis results of the weak points of the connection, determine the optimization goals, including improving the strength of the connection, reducing the failure rate of the connection, optimizing production efficiency, and reducing production costs. According to the existing problems and optimization goals, design an optimization plan for the weak points of the connection, including adjusting the production process, optimizing production parameters, improving connection technology, etc.

[0073] Use mathematical models, computer simulation tools, etc. to simulate and evaluate the optimization scheme of the design to evaluate the effect of the optimization scheme and predict its performance in actual production. According to the results of simulation and evaluation, optimize and adjust the optimization scheme to make it more in line with the expected optimization goal. According to the optimized optimization scheme, generate the first process optimization result, which includes the quantitative evaluation of various optimization indicators and the final production process plan.

[0074] The focus of control optimization under process improvement is to fundamentally solve the problem of weak connection points by improving and optimizing the process flow. Specifically, the process improvement plan is optimized and adjusted using the same optimization method, including modifying parameter settings, adjusting the process flow, selecting suitable connection materials, etc., to obtain the second process optimization result.

[0075] Evaluate the implementation costs of the first process optimization results and the second process optimization results, including production costs, human resource costs, equipment investment costs, etc., and determine which plan is more economical and reasonable by comparing the cost of the two plans; evaluate the implementation effects of the first process optimization results and the second process optimization results, including evaluation of connection quality, production efficiency, product performance, etc., and determine which plan is more effective in practical applications by comparing the effects of the two plans.

[0076] A balance analysis is conducted between cost and effect, including weighing the relationship between the two to find an optimal balance point, that is, to save costs while ensuring quality and performance in terms of effect, in order to determine which solution is better. Based on the results of the balance analysis, the parameters of the first process optimization results and the second process optimization results are optimized, with the goal of reducing costs as much as possible while ensuring quality and performance. Finally, the process solution after parameter optimization is confirmed to achieve the best balance between cost and effect.

[0077] Furthermore, the method further comprises:

[0078] Perform re-sample production test based on parameter optimization results and receive test results;

[0079] Determine whether the test result has the same direction of similarity as the force fitting result;

[0080] If the test result and the force fitting result have the same direction similarity, the same direction optimization feedback is generated;

[0081] Continued optimization management is performed through the same-direction optimization feedback.

[0082] Based on the parameter optimization results, a batch of samples are produced and tested to verify the performance of the process solution after parameter optimization in actual production. The tests may include connection strength tests, structural stability tests, etc., and the test results are received.

[0083] The test results are analyzed and compared with the previous force fitting results. The main purpose is to determine whether the process scheme after parameter optimization can effectively improve the quality of the connection. If the test results and the force fitting results have the same direction of similarity, it means that the process scheme after parameter optimization is successful and can achieve the expected effect; otherwise, it is necessary to further adjust the parameters or process flow.

[0084] If the test results and the force fitting results have the same direction of similarity, it means that the process scheme after parameter optimization has achieved good results in actual production. In this case, the same direction optimization feedback is generated.

[0085] Based on the generated same-direction optimization feedback, continued optimization management is carried out, that is, further improving the process flow, optimizing parameter settings, updating production equipment or materials, etc. according to the previous optimization direction. The goal of optimization management is to continuously improve the production process to ensure that it can adapt to changing production needs and continuously improve production quality.

[0086] In summary, the vehicle body connection process optimization method based on the multi-physics field model provided in the embodiment of the present application has the following technical effects:

[0087] 1. By establishing a virtual digital twin model, the sample data and production parameters in actual production are combined with the mathematical analysis model to achieve comprehensive modeling and analysis of the body connection process, thereby improving the scientific nature of the optimization process;

[0088] 2. The image acquisition and adaptive enhancement processing technology can accurately capture the key features of the body connection process, which provides real-time data support and more accurate and comprehensive information for the optimization process;

[0089] 3. By obtaining the application scenario data set and configuring multiple physical field models based on the data set, multi-faceted optimization of the virtual digital twin model is achieved, which makes the optimization process more comprehensive and can fully consider the impact of different physical scenarios on the connection process;

[0090] 4. The weak points of the connection are constructed through the force fitting results, and these weak points are used to optimize the parameters of the production parameters. This enables the optimization process to improve the weak links of the connection process in a targeted manner and improve its quality and stability.

[0091] In summary, this method effectively solves the limitations of traditional body connection process optimization, improves the scientificity, accuracy and comprehensiveness of the optimization process, and thus achieves the technical effect of improving the quality and stability of the body connection process.

[0092] Based on the same inventive concept as the method for optimizing the vehicle body connection process based on the multi-physics field model in the aforementioned embodiment, Figure 2 As shown, the present application provides a car body connection process optimization system based on a multi-physics field model, the system comprising:

[0093] A sample establishment module 10, wherein the sample establishment module 10 is used to establish a proofing sample, wherein the proofing sample is a connection sample of a vehicle body, and the proofing sample is provided with identifications of production parameters and sample data;

[0094] A model building module 20, wherein the model building module 20 is used to build a virtual digital twin model with the production parameters and the sample data, wherein the virtual digital twin model is an established mathematical analysis model;

[0095] An image acquisition module 30, wherein the image acquisition module 30 is used to acquire images of the proofing sample and establish an image set of the proofing sample, wherein the image set has a position identifier and includes a digital image set and a radiographic image set;

[0096] An image enhancement processing module 40, the image enhancement processing module 40 is used to input the image set into an image processing network, perform adaptive enhancement processing on the image, and extract connection features of the proofing sample;

[0097] A model updating module 50, wherein the model updating module 50 is used to synchronize the connection features to the virtual digital twin model according to the location identifier, so as to update the virtual digital twin model;

[0098] A force fitting module 60, which is used to obtain an application scenario data set, configure M physical field models based on the application scenario data set, perform force fitting of the updated virtual digital twin model with the M physical field models, and establish a force fitting result, wherein M is an integer greater than 1;

[0099] The parameter optimization module 70 is used to construct connection weak points through the force fitting results, and to optimize the parameters of production parameters through the connection weak points.

[0100] Furthermore, the system further includes an enhanced processing result acquisition module to perform the following operation steps:

[0101] Performing image classification of the image set by a preprocessing unit to establish a digital image set;

[0102] Performing distortion correction on the digital image set based on acquisition parameters, and arranging and splicing the distortion correction results according to the position identifiers to generate an arrangement and splicing result;

[0103] The arrangement and splicing result is subjected to perspective recognition, the perspective recognition result is subjected to enhancement processing based on an image enhancement unit, and the enhancement processing result is output.

[0104] Furthermore, the system also includes an arrangement and splicing result generation module to perform the following operation steps:

[0105] Perform initial arrangement and splicing based on the position identifier to establish a basic position splicing result;

[0106] Performing feature alignment search at the same position through the basic position splicing result, wherein the feature alignment search is a process of performing feature search and feature alignment within a preset tolerant anchor frame;

[0107] Based on the feature alignment search, the basic position stitching result is corrected and compensated to generate the arrangement stitching result.

[0108] Furthermore, the system further includes a correction compensation module to perform the following operation steps:

[0109] Set similarity matching threshold;

[0110] For any feature alignment search, if the similarity matching threshold of the feature satisfies the similarity matching threshold, the moving direction and moving distance of the corresponding feature alignment search are recorded as calibration data;

[0111] Arrange all calibration data, establish a calibration data set, and perform alignment correction on feature alignment search that does not meet the similarity matching threshold based on the calibration data set;

[0112] The correction compensation of the basic position stitching result is completed according to the alignment correction result.

[0113] Furthermore, the system further comprises an alignment correction module to perform the following operation steps:

[0114] Performing cluster analysis of movement direction and movement distance based on the calibration data set, and generating a first movement reference vector based on a maximum clustering result;

[0115] Performing positional proximity matching on the feature alignment search that does not meet the similarity matching threshold through the maximum clustering result, and constructing a second movement reference vector according to the positional proximity matching result;

[0116] A reference coefficient of the first moving reference vector and the second moving reference vector is configured, and alignment correction is performed based on the reference coefficient, the first moving reference vector, and the second moving reference vector.

[0117] Furthermore, the system also includes a parameter optimization module to perform the following operation steps:

[0118] Based on the weak connection point, control optimization is performed under the same production process to generate a first process optimization result;

[0119] Based on the weak connection point, control optimization is performed under process improvement to generate a second process optimization result;

[0120] A cost-effective balance analysis is performed on the first process optimization result and the second process optimization result, and parameter optimization is completed based on the balance analysis results.

[0121] Furthermore, the system further includes a continuous optimization management module to perform the following operation steps:

[0122] Perform re-sample production test based on parameter optimization results and receive test results;

[0123] Determine whether the test result has the same direction of similarity as the force fitting result;

[0124] If the test result and the force fitting result have the same direction similarity, the same direction optimization feedback is generated;

[0125] Continued optimization management is performed through the same-direction optimization feedback.

[0126] Through the above-mentioned detailed description of the vehicle body connection process optimization method based on the multi-physical field model in this specification, those skilled in the art can clearly understand the vehicle body connection process optimization system based on the multi-physical field model in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0127] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the processor of the computer device is used to provide computing and control capabilities; the memory of the computer device includes a non-volatile storage medium and an internal memory, wherein the non-volatile storage medium stores an operating system, a computer program and a database, and the internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium; the network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement a body connection process optimization method based on a multi-physics field model.

[0128] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0129] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0130] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. Car body connection process optimization method based on multi-physics field model, It is characterized in that The method comprises: establishing a proofing sample, wherein the proofing sample is a connection sample of a vehicle body, and the proofing sample is provided with identifications of production parameters and sample data, wherein the production parameters include welding temperature, pressure, welding time, and riveting pressure, and the sample data includes connection technology, connection material, connection position, and production parameters; Establishing a virtual digital twin model with the production parameters and the sample data, wherein the virtual digital twin model is an established mathematical analysis model; Capturing images of the proofing samples to establish an image set of the proofing samples, wherein the image set has a position identifier; Inputting the image set into an image processing network, performing adaptive enhancement processing on the image, and extracting connection features of the proofing samples; Synchronizing the connection features to the virtual digital twin model according to the location identifier, thereby updating the virtual digital twin model; Acquire an application scenario data set, configure M physical field models based on the application scenario data set, perform force fitting of the updated virtual digital twin model with the M physical field models, and establish a force fitting result, where M is an integer greater than 1; The connection weak points are constructed through the force fitting results, and the parameters of the production parameters are optimized through the connection weak points.

2. The method according to claim 1, It is characterized in that The step of inputting the image set into an image processing network, performing adaptive enhancement processing on the image, and extracting connection features of the proofing samples comprises: Performing image classification of the image set by a preprocessing unit to establish a digital image set; Performing distortion correction on the digital image set based on acquisition parameters, and arranging and splicing the distortion correction results according to the position identifiers to generate an arrangement and splicing result; The arrangement and splicing result is subjected to perspective recognition, the perspective recognition result is subjected to enhancement processing based on an image enhancement unit, and the enhancement processing result is output.

3. The method according to claim 2, It is characterized in that The step of performing distortion correction on the digital image set based on acquisition parameters, and arranging and splicing the distortion correction results according to the position identifiers to generate an arrangement and splicing result includes: Perform initial arrangement and splicing based on the position identifier to establish a basic position splicing result; Performing feature alignment search at the same position through the basic position splicing result, wherein the feature alignment search is a process of performing feature search and feature alignment within a preset tolerant anchor frame; Based on the feature alignment search, the basic position stitching result is corrected and compensated to generate the arrangement stitching result.

4. The method according to claim 3, It is characterized in that The performing feature alignment search at the same position through the basic position splicing result includes: Set similarity matching threshold; For any feature alignment search, if the similarity matching threshold of the feature satisfies the similarity matching threshold, the moving direction and moving distance of the corresponding feature alignment search are recorded as calibration data; Arrange all calibration data, establish a calibration data set, and perform alignment correction on feature alignment search that does not meet the similarity matching threshold based on the calibration data set; The correction compensation of the basic position stitching result is completed according to the alignment correction result.

5. The method according to claim 4, It is characterized in that The alignment correction of the feature alignment search that does not meet the similarity matching threshold based on the calibration data set also includes: Performing cluster analysis of movement direction and movement distance based on the calibration data set, and generating a first movement reference vector based on a maximum clustering result; Performing positional proximity matching on the feature alignment search that does not meet the similarity matching threshold through the maximum clustering result, and constructing a second movement reference vector according to the positional proximity matching result; A reference coefficient of the first moving reference vector and the second moving reference vector is configured, and alignment correction is performed based on the reference coefficient, the first moving reference vector, and the second moving reference vector.

6. The method according to claim 1, It is characterized in that The parameter optimization of production parameters through the connection weak points includes: Based on the connection weak points, control optimization is performed under the same production process to generate a first process optimization result; Based on the weak connection point, control optimization is performed under process improvement to generate a second process optimization result; A cost-effective balance analysis is performed on the first process optimization result and the second process optimization result, and parameter optimization is completed based on the balance analysis results.

7. The method according to claim 1, It is characterized in that The method further comprises: Perform re-sample production test based on parameter optimization results and receive test results; Determine whether the test result has the same direction of similarity as the force fitting result; If the test result and the force fitting result have the same direction similarity, the same direction optimization feedback is generated; Continued optimization management is performed through the same-direction optimization feedback.

8. Car body connection process optimization system based on multi-physics field model, It is characterized in that A system for implementing the vehicle body connection process optimization method based on a multi-physics field model according to any one of claims 1 to 7, the system comprising: A sample establishment module, the sample establishment module is used to establish a proofing sample, wherein the proofing sample is a connection sample of a vehicle body, and the proofing sample is provided with identifications of production parameters and sample data, the production parameters include welding temperature, pressure, welding time, riveting pressure, and the sample data includes connection technology, connection material, connection part, and production parameters; A model building module, wherein the model building module is used to build a virtual digital twin model with the production parameters and the sample data, wherein the virtual digital twin model is an established mathematical analysis model; An image acquisition module, the image acquisition module is used to acquire images of the proofing sample and establish an image set of the proofing sample, wherein the image set has a position identifier; An image enhancement processing module, the image enhancement processing module is used to input the image set into an image processing network, perform adaptive enhancement processing on the image, and extract connection features of the proofing samples; A model updating module, wherein the model updating module is used to synchronize the connection features to the virtual digital twin model according to the location identifier, so as to update the virtual digital twin model; A force fitting module, the force fitting module is used to obtain an application scenario data set, configure M physical field models based on the application scenario data set, perform force fitting of the updated virtual digital twin model with the M physical field models, and establish a force fitting result, wherein M is an integer greater than 1; A parameter optimization module is used to construct connection weak points through the force fitting results, and to optimize the parameters of production parameters through the connection weak points.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the vehicle body connection process optimization method based on a multi-physics field model according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the vehicle body connection process optimization method based on a multi-physics field model according to any one of claims 1 to 7 are implemented.