Process model generation method and system for new energy sheet metal process

Abstract

The present invention relates to the field of computer-aided design technology, and in particular to a method and system for generating a process model of a new energy sheet metal process. The method comprises the following steps: obtaining and analyzing historical new energy vehicle body manufacturing data to obtain a new energy vehicle component topological structure model; modeling the historical new energy vehicle body manufacturing data to obtain a process flow structure model; calculating the historical new energy vehicle body manufacturing data to obtain sheet metal structure deformation influencing factors; evaluating the new energy vehicle component topological structure model to obtain docking component precision error data; performing component docking precision error process identification based on docking component precision error data and sheet metal structure deformation influencing factors, thereby obtaining precision error process data, and optimizing process parameters, thereby obtaining a sheet metal process flow optimization model. The present invention reduces the precision error of sheet metal processes.

Description

Technical Field

[0001] The present invention relates to the technical field of computer-aided design, and in particular to a process model generation method and system for a new energy sheet metal process. Background Art

[0002] As the world's attention to environmental protection and energy resources continues to intensify, the new energy vehicle industry is developing rapidly as an emerging green transportation solution. New energy vehicles are not only required to have excellent performance and safety, but also to meet strict environmental standards. This forces manufacturers to continuously optimize various processes during the production process to ensure the quality and reliability of the vehicle. In particular, the manufacturing of sheet metal parts, as an important part of new energy vehicles, the optimization of its process flow has a vital impact on the overall product quality. The core processes involved in sheet metal technology include stamping, bending, welding, assembly and other links. Traditional sheet metal process flows are usually based on experience and conventional engineering technology, which is difficult to meet the requirements of new energy vehicles for high precision, high consistency and complex structures of sheet metal parts. In addition, traditional methods are insufficient in dealing with component deformation, which easily leads to error accumulation in the manufacturing process, affecting the assembly accuracy and performance of the final product. Summary of the invention

[0003] Based on this, it is necessary for the present invention to provide a process model generation method and system for new energy sheet metal processes to solve at least one of the above technical problems.

[0004] To achieve the above purpose, a method for generating a process model of a new energy sheet metal process comprises the following steps:

[0005] Step S1: acquiring historical new energy vehicle body manufacturing data, and extracting vehicle body component structural features from the historical new energy vehicle body manufacturing data, thereby obtaining new energy vehicle body component structural data; performing new energy vehicle component topological structure analysis based on the new energy vehicle body component structural data, thereby obtaining a new energy vehicle component topological structure model;

[0006] Step S2: extracting sheet metal process features from historical new energy vehicle body manufacturing data to obtain body manufacturing sheet metal process data, and performing sheet metal process flow structure modeling based on the body manufacturing sheet metal process data to obtain a process flow structure model;

[0007] Step S3: estimating the sheet metal structure deformation of the process flow for the historical new energy vehicle body manufacturing data based on the process flow structure model, thereby obtaining the sheet metal structure deformation data of the process flow; calculating the sheet metal structure deformation influencing factor for the body manufacturing sheet metal process data and the sheet metal structure deformation data of the process flow, thereby obtaining the sheet metal structure deformation influencing factor;

[0008] Step S4: using the process flow structure model to simulate the new energy vehicle component docking process on the new energy vehicle component topology structure model, thereby obtaining new energy vehicle component docking simulation data; performing component docking accuracy error evaluation on the new energy vehicle component docking simulation data, thereby obtaining docking component accuracy error data;

[0009] Step S5: Perform component docking precision error process identification based on the docking component precision error data and the sheet metal structure deformation influencing factors to obtain precision error process data, and optimize process parameters of the process flow structure model based on the precision error process data to obtain a sheet metal process flow optimization model.

[0010] The present invention can establish an accurate component topology model by extracting the structural features of the body parts of new energy vehicles and performing topological structure analysis. This lays the foundation for subsequent process optimization and precision improvement, ensuring that the component design is more in line with actual production requirements. The extracted structural data enables the manufacturing process to be adjusted more scientifically based on historical data, thereby reducing the influence of subjective experience and improving the scientificity and accuracy of design and manufacturing. Through the topological structure model, the structural characteristics of key components can be identified, providing a clear basis for the subsequent sheet metal process feature extraction. Sheet metal process feature extraction and process flow structure modeling can establish a detailed process flow structure model, which helps manufacturers understand and optimize the impact of each process step, thereby improving overall production efficiency and product quality. Through detailed analysis of sheet metal process data, the sources of errors in traditional process flows can be identified and reduced, and the consistency and controllability of the process can be improved. The process flow model with detailed data support can provide a scientific basis for improving the process flow, making the optimization of sheet metal processes more accurate and effective. The deformation estimation based on the process flow structure model can predict the deformation of components in actual production, help manufacturers take measures in advance to adjust and control, and reduce problems in production. By calculating the influencing factors of sheet metal structure deformation, it is possible to identify which factors have the greatest impact on the deformation, so as to make targeted process improvements and improve the dimensional stability and quality of the product. Understanding and controlling the influencing factors of sheet metal structure deformation can help improve the stability and consistency of the production process and reduce uncertainty and errors in production. Through component docking process simulation, the errors that may occur in the docking process can be understood in real time, and the process parameters can be adjusted in time to improve the accuracy of component docking. Accuracy error assessment helps identify and solve problems that may occur in actual production, optimize the process of component docking, and ensure that the quality of the final product meets the standards. Through the accurate assessment of docking accuracy, rework caused by non-compliance of components can be reduced and production efficiency can be improved. According to the analysis of the accuracy error data of docking components and the influencing factors of sheet metal structure deformation, the accuracy errors in the process can be accurately identified to help formulate more effective improvement strategies. Optimizing process parameters according to accuracy error process data can improve the production process, improve the assembly accuracy of components and the overall product quality. The optimized process flow model can effectively reduce errors and defects in production, improve production efficiency, and reduce material waste and production costs. In summary, through the implementation of the above steps, the manufacturing process of sheet metal parts for new energy vehicles can be significantly optimized, improving the consistency of component quality and the stability of the production process.

[0011] Optionally, step S1 specifically includes:

[0012] Step S11: acquiring historical new energy vehicle body manufacturing data, and extracting vehicle body component structural features from the historical new energy vehicle body manufacturing data, thereby obtaining new energy vehicle body component structural data;

[0013] Step S12: extracting structural material features from the structural data of the new energy vehicle body parts to obtain body structure material data, and performing material property analysis based on the body structure material data to obtain body structure material property data;

[0014] Step S13: classifying the body component structure data of the new energy vehicle according to the body component structure connection mode, thereby obtaining component structure welding data and component structure splicing data;

[0015] Step S14: integrating the structural connection relationship of the vehicle body components according to the component structure welding data and the component structure splicing data, thereby obtaining the structural connection data of the vehicle body components;

[0016] Step S15: Performing a topological structure analysis on the body parts based on the body part structure connection data to obtain a body part topological structure model, and integrating the body structure characteristics of the body part topological structure model according to the body structure material characteristic data to obtain a new energy vehicle component topological structure model.

[0017] The present invention can establish a comprehensive body manufacturing data set by acquiring historical data. These data can cover different models, production periods and manufacturing processes. Extracting the structural features of body parts helps to identify and record key parameters such as the shape, size, and configuration of body parts. This provides basic data for subsequent analysis and optimization. By comparing different historical data, the evolution trend of body design can be identified, and then the existing design and manufacturing process can be optimized to improve the safety and durability of the body. Extracting structural material features helps to understand the types and characteristics of materials used in different body parts, including strength, stiffness, corrosion resistance, etc. This is crucial for evaluating the performance and safety of the body. By analyzing the material properties, it is possible to find out which materials have the best performance and apply these materials in the design to improve the overall performance and longevity of the body. Analyzing data can also help discover the potential of new materials, thereby promoting material innovation in the new energy vehicle industry. Classifying different structural connection methods (such as welding, splicing, etc.) helps to understand the manufacturing process and connection quality of the body. This is important for optimizing production processes and improving production efficiency. Clarifying welding and splicing data helps to formulate better quality control standards to ensure the reliability of body parts in production and use. Understanding the pros and cons of different connection methods can help you choose the best connection method during the design stage, improve the structural strength of the car body and reduce production costs. Integrating welding and splicing data to form a comprehensive connection relationship diagram helps analyze the overall structural stability and connection reliability of the car body components. By analyzing the connection relationship data, potential connection problems can be identified and measures can be taken in advance for troubleshooting and maintenance. Optimizing the connection relationship helps improve the accuracy and efficiency of the production process and reduce problems in production. Through topological structure analysis, establishing a structural model of the car body components helps to understand the overall layout and structural strength of the car body. Integrating the topological structure model in combination with material property data can comprehensively evaluate the structural performance of the car body, including its strength, durability and safety. The established topological structure model can be used to optimize the design and improve the structural performance and manufacturing efficiency of the car body. At the same time, simulation tests can be carried out in a virtual environment to predict the performance of the car body in actual use and reduce modifications in actual production.

[0018] Optionally, step S13 is specifically:

[0019] Step S131: classifying the component structure data of the new energy vehicle body to obtain the connection component structure data and the main component structure data;

[0020] Step S132: extracting the main component structure connection structure features from the main component structure data, thereby obtaining the main component connection structure data, and calculating the connection structure similarity between the component connection structure data and the main component connection structure data, thereby obtaining the connection structure similarity data;

[0021] Step S133: integrating the component structure splicing relationship of the connection component structure data and the main component structure data according to the connection structure similarity data, thereby obtaining component structure splicing data;

[0022] Step S134: performing component structure roughness calculation on the main component structure data to obtain component structure roughness data, and performing structure roughness threshold statistics based on the component structure roughness data to obtain a structure roughness threshold;

[0023] Step S135: clustering the main component structure data into high-roughness structures according to the structure roughness threshold, thereby obtaining high-roughness structure data;

[0024] Step S136: Integrate the high-roughness structure connection relationship of the main component structure data according to the high-roughness structure data, so as to obtain component structure welding data.

[0025] By classifying the structural data of the vehicle body components, the present invention can effectively divide the data into connecting components and main components, which helps to more accurately analyze and optimize the relationship between different components. Extracting the structural connection features of the main components and calculating the similarity helps to identify and compare the similarity of component connections, thereby optimizing the connection design and improving the reliability of the overall structure. Integrating the structural splicing relationship of the components according to the similarity data helps to achieve effective splicing and matching between components during the design and manufacturing process, thereby improving production efficiency and component quality. Calculating the roughness of the component structure and performing threshold statistics helps to evaluate the surface quality of the components, ensure that the components meet manufacturing standards, and improve the assembly accuracy and durability of the components. Clustering the high-roughness structure data can identify areas with higher surface roughness, thereby providing a basis for subsequent finishing or processing and improving the flatness of the overall component. Integrating the connection relationship of high-roughness structures can optimize the welding process and component connection design, thereby improving the overall strength and stability of the structure.

[0026] Optionally, step S2 specifically includes:

[0027] Step S21: performing body mode manufacturing process clustering on historical new energy vehicle body manufacturing data to obtain body mode manufacturing process data, and performing sheet metal process flow feature extraction and sheet metal process parameter feature extraction on the body mode manufacturing process data to obtain sheet metal process flow data and sheet metal process parameter data;

[0028] Step S22: performing abnormal sheet metal process parameter identification according to the sheet metal process parameter data, thereby obtaining abnormal sheet metal process parameter data, and performing abnormal sheet metal process parameter correction on the sheet metal process parameter data according to the abnormal sheet metal process parameter data, thereby obtaining sheet metal process correction parameter data;

[0029] Step S23: performing a body mode sheet metal process parameter trend analysis according to the sheet metal process correction parameter data, thereby obtaining body mode sheet metal process parameter trend data;

[0030] Step S24: integrating the process parameters of the sheet metal process data according to the sheet metal process parameter trend data of the vehicle body model, thereby obtaining the sheet metal process data of the vehicle body manufacturing;

[0031] Step S25: Perform sheet metal process flow structure modeling according to the body manufacturing sheet metal process data, so as to obtain a process flow structure model.

[0032] The present invention can classify different body modes into several categories through cluster analysis, making the data organization more structured. This helps to understand the similarities and differences between different body modes. Sheet metal process flow and parameter feature extraction help identify key manufacturing processes and parameters, provide reference for future production, and improve the standardization of production. The obtained sheet metal process flow and parameter data can provide a basis for subsequent process optimization, thereby improving production efficiency and product quality. By identifying abnormal parameters, problems in the production process can be discovered in time to avoid unqualified products. Abnormal parameter correction helps to reduce production rework caused by parameter deviation, saving time and resources. Improve the consistency and stability of process parameters to ensure the stability of production and the reliability of the final product. Trend analysis can predict possible process problems in the future, take measures in advance, and perform preventive maintenance. After identifying the change trend of process parameters, the production process can be adjusted in a targeted manner to optimize the production process. Long-term tracking of process parameter trends helps to continuously improve the manufacturing process and improve overall production efficiency and product quality. The integrated data provides a comprehensive process flow view, making the parameters and processes of each link clear at a glance. Provides a complete data foundation to help management make more accurate decisions. The integrated data helps to find bottlenecks and improvement points in the process, so as to implement effective optimization measures. The established process flow structure model provides a systematic management tool for the production process, which helps to standardize operations. The model can help analyze and optimize the production process and find potential improvement points.

[0033] Optionally, step S22 is specifically:

[0034] Step S221: performing frequency domain conversion on the sheet metal process parameter data to obtain a sheet metal process parameter spectrum;

[0035] Step S222: integrating the frequency amplitude characteristics according to the sheet metal process parameter spectrum, thereby obtaining the sheet metal process frequency amplitude data;

[0036] Step S223: performing frequency amplitude threshold statistics based on the sheet metal process frequency amplitude data, thereby obtaining the sheet metal process frequency amplitude threshold;

[0037] Step S224: clustering the abnormal frequency amplitude spectrum of the sheet metal process parameter spectrum according to the sheet metal process frequency amplitude threshold, so as to obtain the sheet metal process abnormal frequency amplitude spectrum;

[0038] Step S225: performing time domain conversion on the abnormal frequency amplitude spectrum of the sheet metal process, thereby obtaining abnormal sheet metal process parameter data;

[0039] Step S226: performing abnormal sheet metal process parameter correction on the sheet metal process parameter data according to the abnormal sheet metal process parameter data, thereby obtaining sheet metal process correction parameter data.

[0040] The present invention can reveal the frequency characteristics in the data by performing frequency domain conversion on the sheet metal process parameter data, which helps to identify periodic changes and potential sources of problems in the process. By integrating the frequency amplitude characteristics, the key features of the frequency amplitude data can be systematically extracted, thereby making a more comprehensive evaluation of the performance of the sheet metal process. Through frequency amplitude threshold statistics, reasonable standards can be set to help determine which amplitude features are normal and which may be abnormal, thereby providing a benchmark for detecting anomalies. By performing spectrum clustering based on thresholds, abnormal frequency amplitudes can be accurately identified, thereby better locating potential problems or defects in the process. By converting the abnormal frequency amplitude spectrum back to the time domain, the abnormal data performance can be intuitively seen, which helps to further analyze and diagnose the root cause of the problem. By correcting the abnormal data, the overall process conditions affected by the abnormal sheet metal process parameters can be adjusted, the stability and quality of the process can be improved, and the overall production efficiency and product quality can be improved.

[0041] Optionally, step S3 specifically includes:

[0042] Step S31: estimating the sheet metal structure deformation amount of the historical new energy vehicle body manufacturing data, thereby obtaining the sheet metal structure deformation amount data of the body manufacturing;

[0043] Step S32: dividing the sheet metal structure deformation data of the vehicle body manufacturing into sheet metal structure deformation data of the process flow according to the process structure model, thereby obtaining the sheet metal structure deformation data of the process flow;

[0044] Step S33: performing principal component analysis on the sheet metal process data of the vehicle body manufacturing, thereby obtaining principal component combination data of sheet metal process parameters;

[0045] Step S34: constructing a linear regression mathematical model based on the sheet metal structural deformation variable data of the process flow and the principal component combination data of the sheet metal process parameters, thereby obtaining a structural deformation variable regression model;

[0046] Step S35: calculating the sheet metal process parameter regression coefficients for the sheet metal structure deformation variable data of the process flow and the sheet metal process parameter principal component combination data according to the structure deformation variable regression model, thereby obtaining the sheet metal process parameter regression coefficient data;

[0047] Step S36: Select parameters influencing the sheet metal structure deformation variable based on the sheet metal process parameter regression coefficient data, so as to obtain the sheet metal structure deformation variable influencing factors.

[0048] The present invention estimates the deformation data of sheet metal structure in body manufacturing, which helps to understand the deformation problems that may occur in the body during the production process, thereby providing basic data for subsequent optimization design. Dividing the deformation variables into process flow sheet metal structure deformation variables can organize complex deformation data into more meaningful process flow parts, helping to identify the impact of different production stages on deformation. Extracting the principal component combination data of sheet metal process parameters through principal component analysis can reduce the data dimension, reveal the key factors affecting the sheet metal process, and simplify subsequent analysis. Constructing a linear regression mathematical model can reveal the relationship between the deformation variable and the process parameter, and provide a prediction tool for process optimization and quality control. Calculating the regression coefficient of the sheet metal process parameter can quantify the impact of each process parameter on the deformation variable, thereby determining which parameters need to be adjusted to improve the quality of the body. Selecting the influencing factors of the sheet metal structure deformation variable can help optimize the process parameter configuration and ultimately improve the accuracy and reliability of body manufacturing.

[0049] Optionally, step S4 is specifically:

[0050] Step S41: extracting the structure of the spliced ​​components from the topological structure model of the new energy vehicle components, thereby obtaining the spliced ​​structure model of the new energy vehicle components;

[0051] Step S42: performing an automobile component docking process simulation on the new energy automobile component splicing structure model according to the process flow structure model, thereby obtaining new energy automobile component docking simulation data;

[0052] Step S43: performing automobile component similarity calculation according to the new energy automobile component docking simulation data, thereby obtaining automobile component similarity data, and performing similar automobile component docking simulation clustering on the new energy automobile component docking simulation data according to the automobile component similarity data, thereby obtaining new energy automobile component docking simulation clustering data;

[0053] Step S44: performing docking accuracy error calculation according to the new energy vehicle component docking simulation clustering data, thereby obtaining simulated docking accuracy error data;

[0054] Step S45: evaluating the docking component accuracy error of the new energy vehicle component splicing structure model according to the simulated docking accuracy error data, thereby obtaining the docking component accuracy error data.

[0055] The present invention extracts the splicing structure model of new energy vehicle components to clarify the connection relationship between the components, which is convenient for subsequent analysis and optimization of the docking process. The data generated by the docking process simulation can help evaluate potential problems in the component docking process, optimize the process flow, and improve assembly accuracy. Similarity calculation and cluster analysis can identify similar component docking patterns, which helps to discover common docking problems and formulate targeted improvement measures. Calculating the docking accuracy error provides simulated error data in the docking process, reveals the shortcomings of the docking accuracy, and provides a specific basis for improvement. The docking component accuracy error assessment can quantify the error level between components, so as to adjust the design or process in a targeted manner and improve the overall assembly quality and accuracy.

[0056] Optionally, step S45 is specifically:

[0057] Step S451: extracting component accuracy error structural features according to the simulated docking accuracy error data, thereby obtaining component accuracy error structural data;

[0058] Step S452: performing structural accuracy error statistics according to the component accuracy error structural data, thereby obtaining component high-accuracy error structural data and component low-accuracy error structural data;

[0059] Step S453: classifying the high-precision error amount structure data of the component and the low-precision error amount structure data of the component according to the proportion of the high-precision error amount structure, so as to obtain the high-precision error amount structure high proportion component data and the low-precision error amount structure high proportion component data;

[0060] Step S454: performing component structure association on the new energy vehicle component splicing structure model according to the high-precision error amount structure high-ratio component data, thereby obtaining high-precision error docking component data; performing component structure association on the new energy vehicle component splicing structure model according to the low-precision error amount structure high-ratio component data, thereby obtaining low-precision error docking component data;

[0061] Step S455: merging the high-precision error docking component data and the low-precision error docking component data to obtain docking component precision error data.

[0062] The present invention extracts the structural features of component accuracy errors to identify the pattern and distribution of errors, such as whether the errors are concentrated in certain components or areas. This helps to understand the root causes of the errors, provides detailed data support for subsequent accuracy analysis and optimization, and facilitates in-depth analysis of the sources and influencing factors of the errors. The statistical accuracy error amount can distinguish the errors into high-precision errors and low-precision errors. This classification helps to clarify which components or connection points have higher errors, and giving priority to solving these problems can significantly improve the overall assembly quality. Providing quantitative data on the error amount makes it more accurate to evaluate the severity and impact of the errors, which helps to formulate specific improvement measures. By classifying the proportion of high-precision error amount structures and low-precision error amount structures, it is possible to preferentially identify components or structures with a higher proportion of errors. This sorting helps to concentrate resources and energy to solve the most critical error problems. Clarifying the error amount structure with a high proportion can provide targeted improvement directions and help the design and production teams to make effective adjustments and optimizations. Through the structural correlation analysis of the high-precision and low-precision error-proportion component data, the specific components and structures where the errors occur can be more accurately identified and located, thereby providing a strong basis for improving the design and process. This correlation analysis helps to optimize component design and docking processes, reduce precision errors, and improve the matching degree between components, thereby improving the performance and reliability of the overall system. Combining the high-precision and low-precision error docking component data can provide a comprehensive view of docking accuracy errors. This allows for a more comprehensive understanding and evaluation of the errors in the assembly process. The comprehensive error data enables the overall error distribution and impact to be considered when formulating improvement measures, ensuring that the optimization measures can effectively improve the overall assembly accuracy and enhance product quality.

[0063] Optionally, step S5 specifically includes:

[0064] Step S51: extracting high-precision error component structure features according to the docking component precision error data, thereby obtaining high-precision error component structure data;

[0065] Step S52: extracting component structure process flow parameters from the high-precision error component structure data according to the process flow structure model, thereby obtaining error component structure process flow parameter data;

[0066] Step S53: Associating the process error influencing parameters with the process error influencing parameters of the error component structure process parameter data according to the sheet metal structure deformation influencing factors, thereby obtaining the precision error process data;

[0067] Step S54: optimizing the process flow structure model for minimum precision error of process parameters according to the precision error process data, thereby obtaining a sheet metal process flow optimization model.

[0068] The present invention can accurately obtain the actual error data of each component by extracting the precision error structural characteristics of the component. These data can help identify and analyze the error sources in the manufacturing process, and the high-precision error data provides reliable basic data for subsequent process optimization. It ensures that the follow-up is carried out on the basis of accurate errors, thereby improving the overall optimization effect, clarifying the error characteristics of the component, and helping to accurately determine the optimization target and scope, avoiding blind optimization. By extracting the component structure process parameters, the role and performance of each component in the process can be clearly understood. This helps to identify which process steps have the greatest impact on the error, making the analysis of the source of the error more systematic, and can associate the error data with the specific parameters of the process, so as to locate the problem more effectively. Obtaining process parameter data helps to more accurately control the variables in the production process and improve the overall quality and consistency of the product. By performing correlation analysis on the influencing factors of the sheet metal structure deformation variable, it is possible to deeply understand how each process parameter affects the error. This in-depth analysis helps to find the root cause of the error. By understanding the specific performance of the error influencing factors, the process can be adjusted in a targeted manner to reduce the error, thereby improving product accuracy and production efficiency. By optimizing the process parameters for minimum precision error, an optimization model for a specific process can be obtained. This model helps to achieve the lowest error in the production process and improve the quality and consistency of the product. The optimized process can more effectively control and reduce errors, thereby improving production efficiency, reducing scrap rate and rework costs. Reducing errors in the production process not only improves product quality, but also may reduce production costs, thereby improving the economic benefits of the enterprise.

[0069] Optionally, the present invention further provides a process model generation system for a new energy sheet metal process, which is used to execute the process model generation method for a new energy sheet metal process as described above, and the process model generation system for the new energy sheet metal process includes:

[0070] The component topology structure analysis module is used to obtain historical new energy vehicle body manufacturing data, and extract the structural features of the vehicle body components from the historical new energy vehicle body manufacturing data, so as to obtain the structural data of the new energy vehicle body components; perform the topology structure analysis of the new energy vehicle components based on the structural data of the new energy vehicle body components, so as to obtain the topology structure model of the new energy vehicle components;

[0071] The sheet metal process flow structure modeling module is used to extract sheet metal process features from historical new energy vehicle body manufacturing data, thereby obtaining body manufacturing sheet metal process data, and to perform sheet metal process flow structure modeling based on the body manufacturing sheet metal process data, thereby obtaining a process flow structure model;

[0072] The deformation variable influencing factor calculation module is used to estimate the deformation variable of the sheet metal structure of the process flow for the historical new energy vehicle body manufacturing data based on the process flow structure model, so as to obtain the sheet metal structure deformation variable data of the process flow; the sheet metal structure deformation variable influencing factor is calculated for the body manufacturing sheet metal process data and the sheet metal structure deformation variable data of the process flow, so as to obtain the sheet metal structure deformation variable influencing factor;

[0073] The docking accuracy error evaluation module is used to simulate the docking process of new energy vehicle components on the topological structure model of new energy vehicle components through the process flow structure model, so as to obtain the docking simulation data of new energy vehicle components; and to evaluate the docking accuracy error of components on the docking simulation data of new energy vehicle components, so as to obtain the docking component accuracy error data;

[0074] The process model optimization module is used to identify the component docking precision error process according to the docking component precision error data and the influencing factors of the sheet metal structure deformation, so as to obtain the precision error process data, and optimize the process parameters of the process flow structure model according to the precision error process data, so as to obtain the sheet metal process flow optimization model.

[0075] The process model generation system of the new energy sheet metal process of the present invention can realize the process model generation method of any new energy sheet metal process of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the process model generation method of the new energy sheet metal process. The internal modules of the system cooperate with each other to reduce the precision error of the sheet metal process. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Other features, objects and advantages of the present invention will become more apparent from the detailed description of non-limiting embodiments thereof made with reference to the following drawings:

[0077] Figure 1 A schematic flow chart of the steps of the process model generation method of the new energy sheet metal process of the present invention;

[0078] Figure 2 Detailed step flow diagram of step S1 in the present invention;

[0079] Figure 3 Detailed step flow diagram of step S2 in the present invention;

[0080] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0081] The technical method of the present invention is described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by technicians in this field without creative work are within the scope of protection of the present invention.

[0082] In addition, the accompanying drawings are only schematic illustrations of the present invention and are not necessarily drawn to scale. The same reference numerals in the figures represent the same or similar parts, and their repeated description will be omitted. Some of the block diagrams shown in the accompanying drawings are functional entities and do not necessarily correspond to physically or logically independent entities. The functional entities can be implemented in software form, or implemented in one or more hardware modules or integrated circuits, or implemented in different networks and / or processor methods and / or microcontroller methods.

[0083] It should be understood that, although the terms "first", "second", etc. may be used herein to describe various units, these units should not be limited by these terms. These terms are used only to distinguish one unit from another unit. For example, without departing from the scope of the exemplary embodiments, the first unit may be referred to as the second unit, and similarly the second unit may be referred to as the first unit. The term "and / or" used herein includes any and all combinations of one or more of the listed associated items.

[0084] To achieve this, please refer to Figures 1 to 3 The present invention provides a method for generating a process model of a new energy sheet metal process, the method comprising the following steps:

[0085] Step S1: acquiring historical new energy vehicle body manufacturing data, and extracting vehicle body component structural features from the historical new energy vehicle body manufacturing data, thereby obtaining new energy vehicle body component structural data; performing new energy vehicle component topological structure analysis based on the new energy vehicle body component structural data, thereby obtaining a new energy vehicle component topological structure model;

[0086] In this embodiment, the geometric dimensions, material properties and assembly method of the body parts of the new energy vehicle are obtained through the new energy vehicle manufacturing workshop. The geometric features of the body parts, such as length, width and thickness, are extracted through computer vision technology, and the mechanical properties of the parts are modeled using the finite element analysis method. The topological structure analysis includes using graph theory methods to identify the connection relationship and assembly sequence between the body parts, and finally establishes the topological structure model of the parts, such as a network structure represented by nodes and edges.

[0087] Step S2: extracting sheet metal process features from historical new energy vehicle body manufacturing data to obtain body manufacturing sheet metal process data, and performing sheet metal process flow structure modeling based on the body manufacturing sheet metal process data to obtain a process flow structure model;

[0088] In this embodiment, sheet metal process features of body manufacturing, including sheet metal thickness, bending radius, stamping process parameters, etc., are extracted from historical new energy vehicle body manufacturing data. These feature data are used to build a sheet metal process flow structure model. For example, a flow chart is used to describe the process steps and their sequence such as stamping, bending, and welding. The process parameters of different process steps, such as stamping pressure and temperature, are considered during modeling.

[0089] Step S3: estimating the sheet metal structure deformation of the process flow for the historical new energy vehicle body manufacturing data based on the process flow structure model, thereby obtaining the sheet metal structure deformation data of the process flow; calculating the sheet metal structure deformation influencing factor for the body manufacturing sheet metal process data and the sheet metal structure deformation data of the process flow, thereby obtaining the sheet metal structure deformation influencing factor;

[0090] In this embodiment, the deformation of the sheet metal parts in the body manufacturing data is estimated based on the process flow structure model. Finite element analysis is used to simulate the deformation of the sheet metal during the process, such as the shape change of the sheet metal after stamping. The influencing factors of the sheet metal structure deformation, such as the influence of temperature change on the deformation, are calculated and compared with the actual deformation data to identify the main influencing factors of the deformation.

[0091] Step S4: using the process flow structure model to simulate the new energy vehicle component docking process on the new energy vehicle component topology structure model, thereby obtaining new energy vehicle component docking simulation data; performing component docking accuracy error evaluation on the new energy vehicle component docking simulation data, thereby obtaining docking component accuracy error data;

[0092] In this embodiment, the process flow structure model is used to simulate the docking process of new energy vehicle components. The simulation includes the positioning, docking and welding process of the components, and generates docking simulation data. The actual docking accuracy measurement tool is used to evaluate the accuracy error of the simulation data and identify the error range, such as docking gap and misalignment.

[0093] Step S5: Perform component docking precision error process identification based on the docking component precision error data and the sheet metal structure deformation influencing factors to obtain precision error process data, and optimize process parameters of the process flow structure model based on the precision error process data to obtain a sheet metal process flow optimization model.

[0094] In this embodiment, the precision error data of the docking parts is combined with the influencing factors of the sheet metal structure deformation to identify the root cause of the precision error process. These data are used to optimize the process flow structure model and adjust the process parameters, such as adjusting the welding temperature and pressure, so as to establish a sheet metal process flow optimization model to improve the precision of the parts docking and the overall body quality.

[0095] Optionally, step S1 specifically includes:

[0096] Step S11: acquiring historical new energy vehicle body manufacturing data, and extracting vehicle body component structural features from the historical new energy vehicle body manufacturing data, thereby obtaining new energy vehicle body component structural data;

[0097] In this embodiment, the manufacturing data of the new energy vehicle body including the geometric shape, size, and process layout of each body component are extracted from the historical data of the new energy vehicle manufacturing workshop. For example, computer vision technology or CAD model is used to analyze these data to extract the structural features of the body components, such as the specific size and position of the crossbeam, A-pillar, B-pillar, etc. For example, if the front crossbeam of a certain model is 70 cm wide and 3 cm thick, these data will be recorded as structural feature data.

[0098] Step S12: extracting structural material features from the structural data of the new energy vehicle body parts to obtain body structure material data, and performing material property analysis based on the body structure material data to obtain body structure material property data;

[0099] In this embodiment, the structural material data of the new energy vehicle body parts are analyzed to extract the material type (such as high-strength steel, aluminum alloy, etc.), thickness, density and other characteristics. The material properties are analyzed through material tests (such as tensile tests, hardness tests), and the material strength, toughness, ductility and other data are obtained. For example, the tensile strength of high-strength steel is 600 MPa and the thickness is 2 mm. These data will be used for structural material property analysis.

[0100] Step S13: classifying the body component structure data of the new energy vehicle according to the body component structure connection mode, thereby obtaining component structure welding data and component structure splicing data;

[0101] In this embodiment, the connection methods of the body component structure are classified to identify different connection methods such as welding, bolt connection, and riveting. Using the data of the connection method, a welding data set (such as weld length, weld point location) and a splicing data set (such as splicing edge, overlapping area) are created. For example, the A-pillar of the body is laser welded, and the weld length is 50 cm, which will be recorded as welding data.

[0102] Step S14: integrating the structural connection relationship of the vehicle body components according to the component structure welding data and the component structure splicing data, thereby obtaining the structural connection data of the vehicle body components;

[0103] In this embodiment, the welding data and splicing data of the component structure are integrated to analyze the connection relationship between the body components. The connection map of the body components is constructed by combining the welding and splicing data. For example, the front cross beam is connected to the A-pillar by welding and connected to the body side panel by riveting. These data will form the body component structure connection data.

[0104] Step S15: Performing a topological structure analysis on the body parts based on the body part structure connection data to obtain a body part topological structure model, and integrating the body structure characteristics of the body part topological structure model according to the body structure material characteristic data to obtain a new energy vehicle component topological structure model.

[0105] In this embodiment, a topological structure analysis is performed based on the structural connection data of the body parts to establish a topological structure model of the body parts. Combined with material property data, such as the distribution of high-strength steel and aluminum alloy, a complete topological structure model is integrated. For example, the topological structure model of the body parts shows that the strength area of ​​the front of the body is concentrated at the A-pillar and the front crossbeam. These data will form the final topological structure model of the new energy vehicle parts.

[0106] Optionally, step S13 is specifically:

[0107] Step S131: classifying the component structure data of the new energy vehicle body to obtain the connection component structure data and the main component structure data;

[0108] In this embodiment, the three-dimensional data of all components are extracted from the body CAD model of the new energy vehicle. These components may include body frames, door panels, front fenders, etc. Using machine learning algorithms (such as cluster analysis), these components are classified according to their geometric shapes and functional characteristics. Component structure classification is based on the shape characteristics (such as rectangles, circles), functions (such as supports, covers) or materials (such as steel, aluminum alloys) of the components. For example, all body frame components are classified as "connection components", and door panels and fenders are classified as "main components". Thereby, the structural data of the connection components and the structural data of the main components are obtained.

[0109] Step S132: extracting the main component structure connection structure features from the main component structure data, thereby obtaining the main component connection structure data, and calculating the connection structure similarity between the component connection structure data and the main component connection structure data, thereby obtaining the connection structure similarity data;

[0110] In this embodiment, structural connection features are extracted from the main component data. Assuming that the main components include a door and a roof, the connection structure of these components may involve welding points, bolt holes, etc. The geometric shape of the connection structure is modeled using feature extraction techniques, such as edge detection and image segmentation. At the same time, the connection structure feature data is generated by calculating the connection point position and angle between the main components. Then, a similarity calculation algorithm (such as Euclidean distance or cosine similarity) is applied to compare the component connection structure data with the main component connection structure data, thereby obtaining the connection structure similarity data. For example, the similarity between the door and the roof in the position and angle of the welding point is calculated.

[0111] Step S133: integrating the component structure splicing relationship of the connection component structure data and the main component structure data according to the connection structure similarity data, thereby obtaining component structure splicing data;

[0112] In this embodiment, the obtained connection structure similarity data is used to integrate the component structure splicing relationship. Assuming that the connection structure similarity data indicates that the door and the body frame have a high degree of similarity, the system will automatically infer that these components may use similar splicing methods (such as welding or bolting). These splicing relationships are integrated into a splicing relationship diagram, including the splicing method and position of the door and the body frame in the example. Finally, the component structure splicing data is obtained, which is used to guide the production and assembly process.

[0113] Step S134: performing component structure roughness calculation on the main component structure data to obtain component structure roughness data, and performing structure roughness threshold statistics based on the component structure roughness data to obtain a structure roughness threshold;

[0114] In this embodiment, the surface roughness of the main component (such as the door panel) is calculated. The surface of the component is scanned with high precision using laser scanning technology to obtain surface point cloud data. The roughness value of the component is calculated using a surface roughness measurement algorithm (such as Ra, Rz, etc.). According to the calculation results, the roughness of the component is statistically analyzed by thresholds to determine the allowable range of the roughness. For example, the roughness threshold of the door panel is set to Ra = 1.5 μm, and the data with a calculated roughness value higher than this threshold will be marked as a high-roughness structure.

[0115] Step S135: clustering the main component structure data into high-roughness structures according to the structure roughness threshold, thereby obtaining high-roughness structure data;

[0116] In this embodiment, based on the obtained roughness data and threshold, the main component data is clustered for high roughness structures. For example, through the K-means clustering algorithm, components with surface roughness higher than Ra=1.5μm are marked as high roughness structures. Assuming that the roughness of some areas of the door panel exceeds the threshold due to welding, these areas will be classified as high roughness structure data.

[0117] Step S136: Integrate the high-roughness structure connection relationship of the main component structure data according to the high-roughness structure data, so as to obtain component structure welding data.

[0118] In this embodiment, the connection relationship of the high roughness structure of the main parts is integrated according to the obtained high roughness structure data. Assuming that the high roughness structure is mainly concentrated in the connection area of ​​the door, the specific data of the connection method (such as welding) between the door and the body frame are integrated through the analysis of these areas. This includes detailed modeling of the welding points in the high roughness area and generating component structure welding data. These data will be used to optimize the production process to ensure that the processing of the high roughness area meets the quality standards.

[0119] Optionally, step S2 specifically includes:

[0120] Step S21: performing body mode manufacturing process clustering on historical new energy vehicle body manufacturing data to obtain body mode manufacturing process data, and performing sheet metal process flow feature extraction and sheet metal process parameter feature extraction on the body mode manufacturing process data to obtain sheet metal process flow data and sheet metal process parameter data;

[0121] In this embodiment, the body manufacturing data of historical new energy vehicles is clustered for body mode manufacturing process analysis. The body manufacturing data is divided into different body mode categories through clustering algorithms (such as K-means or DBSCAN). Then, the sheet metal process flow and sheet metal process parameter feature extraction is performed on the body mode data of each category, and the key sheet metal process flow features and parameter features are extracted using feature selection techniques (such as principal component analysis), and finally the sheet metal process flow data and sheet metal process parameter data are formed. For example, the forming process steps, material thickness, stamping pressure and other data of each body mode are extracted.

[0122] Step S22: performing abnormal sheet metal process parameter identification according to the sheet metal process parameter data, thereby obtaining abnormal sheet metal process parameter data, and performing abnormal sheet metal process parameter correction on the sheet metal process parameter data according to the abnormal sheet metal process parameter data, thereby obtaining sheet metal process correction parameter data;

[0123] In this embodiment, sheet metal process parameter data is used to identify abnormal sheet metal process parameters. Abnormal sheet metal process parameters are identified through statistical analysis (such as Z-score) or machine learning methods (such as anomaly detection algorithms). The identified abnormal parameters are corrected, and correction algorithms (such as regression model correction) are applied to adjust the abnormal parameters to ensure data accuracy. For example, if the stamping thickness of a certain body part exceeds the standard abnormally, the thickness is adjusted to the normal range to obtain sheet metal process correction parameter data.

[0124] Step S23: performing a body mode sheet metal process parameter trend analysis according to the sheet metal process correction parameter data, thereby obtaining body mode sheet metal process parameter trend data;

[0125] In this embodiment, trend analysis is performed based on sheet metal process correction parameter data. Time series analysis (such as ARIMA model) or data visualization tools are used to analyze the changing trend of sheet metal process parameters of the body mode over time, and the sheet metal process parameter trend data of the body mode is generated. For example, the stamping pressure changing trend of different body modes in different production cycles is analyzed.

[0126] Step S24: integrating the process parameters of the sheet metal process data according to the sheet metal process parameter trend data of the vehicle body model, thereby obtaining the sheet metal process data of the vehicle body manufacturing;

[0127] In this embodiment, the sheet metal process flow data is integrated using the trend data of the sheet metal process parameters of the body mode. Data fusion technology (such as weighted average or data fusion algorithm) is used to integrate the process flow data of different body modes to form comprehensive body manufacturing sheet metal process data. For example, different process parameters obtained in the trend analysis are weighted and integrated to generate the final sheet metal process flow data set.

[0128] Step S25: Perform sheet metal process flow structure modeling according to the body manufacturing sheet metal process data, so as to obtain a process flow structure model.

[0129] In this embodiment, a process flow structure model is established based on the sheet metal process data of the body manufacturing. The process flow structure model is constructed using modeling technology (such as process modeling tools or graph theory methods), and the process steps and their relationships are defined to form a detailed process flow structure model. For example, a structure diagram of the sheet metal process flow is drawn using UML (Unified Modeling Language) or a flowchart tool to represent the input and output of each step and the connection relationship between them.

[0130] Optionally, step S22 is specifically:

[0131] Step S221: performing frequency domain conversion on the sheet metal process parameter data to obtain a sheet metal process parameter spectrum;

[0132] In this embodiment, the original process parameters (such as vibration, pressure, temperature, etc.) collected during the sheet metal processing are converted into the frequency domain through fast Fourier transform (FFT). Assuming that the original data is a time series, collected 1000 times per second, FFT converts these data into a spectrum to show the amplitude of different frequency components. Specifically, the collected signal data is analyzed using the FFT algorithm to obtain a spectrum diagram, in which the X-axis represents the frequency and the Y-axis represents the amplitude.

[0133] Step S222: integrating the frequency amplitude characteristics according to the sheet metal process parameter spectrum, thereby obtaining the sheet metal process frequency amplitude data;

[0134] In this embodiment, the amplitude information of the frequency components is extracted from the obtained spectrum. Assuming that the spectrum shows the amplitudes of different frequencies, the frequency amplitude data is obtained by calculating the total amplitude of each frequency range and integrating them. For example, the frequency range is divided into low, medium, and high frequency bands, and the amplitude value of each frequency band is calculated to form a feature vector containing the amplitude data of these frequency bands.

[0135] Step S223: performing frequency amplitude threshold statistics based on the sheet metal process frequency amplitude data, thereby obtaining the sheet metal process frequency amplitude threshold;

[0136] In this embodiment, the obtained frequency amplitude data is statistically analyzed to determine the amplitude threshold of each frequency segment. For example, a threshold standard is set, the average value and standard deviation of the frequency amplitude are calculated based on historical normal data, and a threshold higher than the mean plus two times the standard deviation is set as the abnormality detection standard. In this way, the amplitude threshold of each frequency segment can be determined.

[0137] Step S224: clustering the abnormal frequency amplitude spectrum of the sheet metal process parameter spectrum according to the sheet metal process frequency amplitude threshold, so as to obtain the sheet metal process abnormal frequency amplitude spectrum;

[0138] In this embodiment, cluster analysis is performed on the obtained spectrum data according to the determined frequency amplitude threshold. The frequency amplitude data is divided into normal and abnormal spectrum classes using the K-means clustering algorithm. For example, if the amplitude of a certain frequency segment exceeds the threshold, it is marked as abnormal, and these abnormal frequency amplitude data are aggregated into an abnormal spectrum graph.

[0139] Step S225: performing time domain conversion on the abnormal frequency amplitude spectrum of the sheet metal process, thereby obtaining abnormal sheet metal process parameter data;

[0140] In this embodiment, the obtained abnormal frequency amplitude spectrum is subjected to inverse FFT conversion to restore it to a time domain signal. Assuming that the abnormal frequency amplitude spectrum shows abnormalities in certain frequency ranges, these spectrum data are converted back to time domain signals using inverse FFT to obtain abnormal sheet metal process parameter data. These data reflect the abnormal signal characteristics within a specific time.

[0141] Step S226: performing abnormal sheet metal process parameter correction on the sheet metal process parameter data according to the abnormal sheet metal process parameter data, thereby obtaining sheet metal process correction parameter data.

[0142] In this embodiment, the abnormal sheet metal process parameter data is used to correct the original sheet metal process parameter data. For example, by comparing the original data with the abnormal data, a data correction algorithm (such as mean correction or weighted average) is applied to adjust the process parameters to eliminate the influence of the abnormality, thereby obtaining the corrected sheet metal process parameter data. These correction data are used to optimize the subsequent process flow.

[0143] Optionally, step S3 specifically includes:

[0144] Step S31: estimating the sheet metal structure deformation amount of the historical new energy vehicle body manufacturing data, thereby obtaining the sheet metal structure deformation amount data of the body manufacturing;

[0145] In this embodiment, in the historical new energy vehicle body manufacturing process, the sheet metal structure is first measured by using a high-precision laser scanner to obtain the actual geometric shape data of each part of the body. Then, the measured actual geometry is compared with the original design model using a computer-aided design (CAD) model. Through difference analysis, the deformation of the sheet metal structure is estimated using finite element analysis (FEA) software. This process can identify the structural deformation generated during the manufacturing process, such as the deformation of the side of the body, the front and rear bumper areas, and generate a specific deformation data set. These data will be used for subsequent process analysis and optimization.

[0146] Step S32: dividing the sheet metal structure deformation data of the vehicle body manufacturing into sheet metal structure deformation data of the process flow according to the process structure model, thereby obtaining the sheet metal structure deformation data of the process flow;

[0147] In this embodiment, based on the obtained sheet metal structure deformation data, the body sheet metal structure is divided according to different stages of the process flow. For example, the data is divided into stamping stage, welding stage and painting stage. Specifically, the deformation of each stage is classified using the process flow structure model. For the stamping stage, the deformation data during the stamping process is analyzed to extract the deformation of specific areas such as the roof or door panels; for the welding stage, the overall deformation of the body after welding is analyzed; for the painting stage, the additional deformation that may be introduced during the painting process is considered. Finally, these data are summarized to obtain detailed process flow sheet metal structure deformation data.

[0148] Step S33: performing principal component analysis on the sheet metal process data of the vehicle body manufacturing, thereby obtaining principal component combination data of sheet metal process parameters;

[0149] In this embodiment, the divided sheet metal process data is subjected to principal component analysis (PCA). First, the data of each process stage is standardized to eliminate dimensional differences. Then, the PCA algorithm is used to extract the main components, which explain the largest variance in the sheet metal process data. Assuming that the data includes parameters such as punching force, temperature, and pressure, the principal component combination data is obtained by PCA calculation, such as principal component 1 (combination of punching force and temperature) and principal component 2 (combination of pressure and temperature). These principal components will be used to build a regression model to simplify complex process data.

[0150] Step S34: constructing a linear regression mathematical model based on the sheet metal structural deformation variable data of the process flow and the principal component combination data of the sheet metal process parameters, thereby obtaining a structural deformation variable regression model;

[0151] In this embodiment, a linear regression model is constructed based on the process flow sheet metal structure deformation variable data and the principal component combination data. Assuming that principal component 1 and principal component 2 are selected as independent variables and the body deformation variable is selected as the dependent variable, a mathematical model is established using linear regression analysis. The regression model is in the form of: deformation variable = β0 + β1 * principal component 1 + β2 * principal component 2 + ε, where β0 is the intercept, β1 and β2 are regression coefficients, and ε is the error term. Through regression analysis, the degree of influence of each principal component on the sheet metal structure deformation variable is determined, and a structural deformation variable regression model is established.

[0152] Step S35: calculating the sheet metal process parameter regression coefficients for the sheet metal structure deformation variable data of the process flow and the sheet metal process parameter principal component combination data according to the structure deformation variable regression model, thereby obtaining the sheet metal process parameter regression coefficient data;

[0153] In this embodiment, the constructed linear regression model is used to calculate the regression coefficients of the process flow sheet metal structure deformation variable data and the sheet metal process parameter principal component combination data. Specifically, the actual data is substituted and the regression coefficient is calculated using the regression model. For example, if the regression coefficient of principal component 1 is 0.5 and the regression coefficient of principal component 2 is 0.3, then these coefficients will show the degree of influence of the sheet metal process parameters on the body deformation variable. The regression coefficient data will help identify the process parameters that have a significant impact on the sheet metal structure deformation variable.

[0154] Step S36: Select parameters influencing the sheet metal structure deformation variable based on the sheet metal process parameter regression coefficient data, so as to obtain the sheet metal structure deformation variable influencing factors.

[0155] In this embodiment, the selection of influencing factors of sheet metal structure deformation variables is performed based on the obtained sheet metal process parameter regression coefficient data. By analyzing the regression coefficient, the parameters with the greatest impact on the deformation variables are identified. For example, if the regression coefficients of certain process parameters are significantly higher than other parameters, it can be inferred that these parameters have a greater impact on the body deformation variables. Assuming that the regression coefficients of stamping temperature and pressure are significantly higher than other parameters, these two parameters are selected as influencing factors. Finally, the main influencing factors of sheet metal structure deformation variables are summarized and confirmed to optimize and improve the manufacturing process.

[0156] Optionally, step S4 is specifically:

[0157] Step S41: extracting the structure of the spliced ​​components from the topological structure model of the new energy vehicle components, thereby obtaining the spliced ​​structure model of the new energy vehicle components;

[0158] In this embodiment, when extracting the splicing component structure of the topological structure model of the new energy vehicle components, first extract the geometric features and interface features of each component from the CAD model, and then align and match these features through an algorithm. For example, the interface surfaces of different components are spliced ​​together using a feature matching algorithm to generate a complete component splicing structure model. In this process, ensure that the contact surfaces and connection points of each component are accurately aligned to prevent subsequent process problems caused by inaccurate splicing.

[0159] Step S42: performing an automobile component docking process simulation on the new energy automobile component splicing structure model according to the process flow structure model, thereby obtaining new energy automobile component docking simulation data;

[0160] In this embodiment, when the docking process simulation is performed on the new energy vehicle component splicing structure model according to the process flow structure model, the various steps and parameters in the process flow are first input into the simulation software. Through simulation, the displacement, mechanical response and possible assembly problems of the components in the actual docking process are calculated. For example, the docking process under different temperature and pressure conditions can be simulated to generate corresponding docking simulation data for subsequent analysis.

[0161] Step S43: performing automobile component similarity calculation according to the new energy automobile component docking simulation data, thereby obtaining automobile component similarity data, and performing similar automobile component docking simulation clustering on the new energy automobile component docking simulation data according to the automobile component similarity data, thereby obtaining new energy automobile component docking simulation clustering data;

[0162] In this embodiment, when calculating the similarity of automobile parts based on the new energy automobile parts docking simulation data, firstly, a set of features, such as geometric shape, interface form, material properties, etc., are selected, and the similarity between the parts is determined by calculating the similarity index of these features (such as Euclidean distance or cosine similarity). Then, a clustering algorithm (such as K-means or hierarchical clustering) is used to process the similarity data, and similar parts docking simulation data are classified into the same category to form new energy automobile parts docking simulation clustering data.

[0163] Step S44: performing docking accuracy error calculation according to the new energy vehicle component docking simulation clustering data, thereby obtaining simulated docking accuracy error data;

[0164] In this embodiment, when calculating the docking accuracy error, the deviation between the actual position of the component and the theoretical position during the actual docking process is first extracted from the simulation data, where the theoretical position can be obtained by three-dimensionally scanning the component with a high-precision measuring device (such as a three-dimensional coordinate measuring machine or a laser scanner), and the actual geometric data of the component is obtained as the theoretical geometric position. The error of each docking point is calculated using an error analysis algorithm (such as the least squares method), and then these errors are statistically analyzed to evaluate the overall docking accuracy. This process can represent the simulated docking accuracy error data through an error graph or a statistical indicator (such as the root mean square error).

[0165] Step S45: evaluating the docking component accuracy error of the new energy vehicle component splicing structure model according to the simulated docking accuracy error data, thereby obtaining the docking component accuracy error data.

[0166] In this embodiment, when the precision error of the splicing structure model of the new energy vehicle components is evaluated based on the simulated docking precision error data, the error data is first compared with the design tolerance of the component. An error analysis tool is used to identify and quantify the sources of errors in the splicing structure and to evaluate the impact of the errors. For example, an error heat map can be generated using precision evaluation software, and optimization suggestions can be made based on these data to improve the overall precision of the docking components.

[0167] Optionally, step S45 is specifically:

[0168] Step S451: extracting component accuracy error structural features according to the simulated docking accuracy error data, thereby obtaining component accuracy error structural data;

[0169] In this embodiment, during the simulation of the docking accuracy error, the component is first scanned in three dimensions by a high-precision measuring device (such as a three-dimensional coordinate measuring machine or a laser scanner) to obtain the actual geometric data of the component. Then, the actual data is compared with the design data to calculate the accuracy error of the component. Using the error data, statistical methods (such as principal component analysis or error distribution analysis) are used to extract the accuracy error structural characteristics of the component. These characteristics include error types (such as dimensional errors, shape errors) and their distribution patterns, which provide a basis for subsequent error statistics and structural analysis.

[0170] Step S452: performing structural accuracy error statistics according to the component accuracy error structural data, thereby obtaining component high-accuracy error structural data and component low-accuracy error structural data;

[0171] In this embodiment, after obtaining the precision error structure data of the component, the error is quantitatively counted through a data analysis tool (such as a data analysis library of MATLAB or Python). Specifically, the errors are classified (such as high-precision errors and low-precision errors) and the magnitude of each error is calculated. For each component, the high-precision error amount (such as an error less than 0.01 mm) and the low-precision error amount (such as an error greater than 0.01 mm but less than 0.1 mm) are counted. The statistical results will generate high-precision error amount structure data and low-precision error amount structure data of the component for subsequent error structure classification and analysis.

[0172] Step S453: classifying the high-precision error amount structure data of the component and the low-precision error amount structure data of the component according to the proportion of the high-precision error amount structure, so as to obtain the high-precision error amount structure high proportion component data and the low-precision error amount structure high proportion component data;

[0173] In this embodiment, the obtained high-precision and low-precision error amount structure data are classified according to the proportion of the error amount to the total error amount of the component. Cluster analysis methods (such as K-means clustering or hierarchical clustering) are used to classify the error amount, and identify components with a high proportion of high-precision error amounts (such as errors accounting for more than 90% of the total error) and components with a high proportion of low-precision error amounts (such as errors accounting for more than 60% of the total error). The classification results generate high-precision error amount structure high-proportion component data and low-precision error amount structure high-proportion component data, providing a basis for component structure association.

[0174] Step S454: performing component structure association on the new energy vehicle component splicing structure model according to the high-precision error amount structure high-ratio component data, thereby obtaining high-precision error docking component data; performing component structure association on the new energy vehicle component splicing structure model according to the low-precision error amount structure high-ratio component data, thereby obtaining low-precision error docking component data;

[0175] In this embodiment, based on the classification results, the splicing structure model of new energy vehicle components is subjected to correlation analysis. For components with a high proportion of high-precision error structures, the component structures associated with them in the splicing structure (such as connection points and load-bearing parts) are analyzed, and the associated component structure information is extracted to obtain high-precision error docking component data. For components with a high proportion of low-precision error structures, the component structures associated with them in the splicing structure (such as connection points and load-bearing parts) are analyzed, and the associated component structure information is extracted to obtain low-precision error docking component data.

[0176] Step S455: merging the high-precision error docking component data and the low-precision error docking component data to obtain docking component precision error data.

[0177] In this embodiment, the obtained high-precision error docking component data and low-precision error docking component data are merged. The high-precision and low-precision error data are merged into a comprehensive data set using a data processing tool (such as Excel or dedicated data processing software). Then, data fusion technology (such as data cleaning and data integration) is applied to process the merged data to ensure the consistency and integrity of the data. Finally, comprehensive docking component accuracy error data is generated for further analysis and optimization of the docking accuracy of the components.

[0178] Optionally, step S5 specifically includes:

[0179] Step S51: extracting high-precision error component structure features according to the docking component precision error data, thereby obtaining high-precision error component structure data;

[0180] In this embodiment, components with high precision errors in the precision error data of the docking components are analyzed to extract structural features of the components such as protrusions, grooves and irregularities of edges.

[0181] Step S52: extracting component structure process flow parameters from the high-precision error component structure data according to the process flow structure model, thereby obtaining error component structure process flow parameter data;

[0182] In this embodiment, when analyzing the process flow structure model of the sheet metal component, a detailed process flow chart is first established, including process steps such as cutting, bending, and welding. The extracted high-precision error component structure data is input using software tools (such as CAD / CAE systems), and the parameters of each process step are extracted, such as bending angle, cutting path, etc. It is ensured that the extracted process flow parameter data can accurately describe the process characteristics of the error component during processing.

[0183] Step S53: Associating the process error influencing parameters with the process error influencing parameters of the error component structure process parameter data according to the sheet metal structure deformation influencing factors, thereby obtaining the precision error process data;

[0184] In this embodiment, based on the influence factors of the deformation of the sheet metal structure, the error sources that may be generated by the error component during the processing are analyzed. For example, the thermal expansion coefficient of the material, possible deformation during the processing, and other factors are considered. By establishing a mathematical model to associate these influencing factors with the process parameters, the influence of each parameter on the error is calculated. Finite element analysis software can be used to simulate and evaluate the contribution of different influencing factors to the final error.

[0185] Step S54: optimizing the process flow structure model for minimum precision error of process parameters according to the precision error process data, thereby obtaining a sheet metal process flow optimization model.

[0186] In this embodiment, the obtained precision error process data is input into the optimization algorithm to optimize the minimum precision error of the process parameters. For example, a genetic algorithm or a particle swarm optimization algorithm is used to adjust the parameters in the process flow model to seek a solution that minimizes the error in all process steps. Ultimately, the sheet metal process flow optimization model obtained should be able to minimize the error and improve the overall processing accuracy of the parts.

[0187] Optionally, the present invention further provides a process model generation system for a new energy sheet metal process, which is used to execute the process model generation method for a new energy sheet metal process as described above, and the process model generation system for the new energy sheet metal process includes:

[0188] The component topology structure analysis module is used to obtain historical new energy vehicle body manufacturing data, and extract the structural features of the vehicle body components from the historical new energy vehicle body manufacturing data, so as to obtain the structural data of the new energy vehicle body components; perform the topology structure analysis of the new energy vehicle components based on the structural data of the new energy vehicle body components, so as to obtain the topology structure model of the new energy vehicle components;

[0189] The sheet metal process flow structure modeling module is used to extract sheet metal process features from historical new energy vehicle body manufacturing data, thereby obtaining body manufacturing sheet metal process data, and to perform sheet metal process flow structure modeling based on the body manufacturing sheet metal process data, thereby obtaining a process flow structure model;

[0190] The deformation variable influencing factor calculation module is used to estimate the deformation variable of the sheet metal structure of the process flow for the historical new energy vehicle body manufacturing data based on the process flow structure model, so as to obtain the sheet metal structure deformation variable data of the process flow; the sheet metal structure deformation variable influencing factor is calculated for the body manufacturing sheet metal process data and the sheet metal structure deformation variable data of the process flow, so as to obtain the sheet metal structure deformation variable influencing factor;

[0191] The docking accuracy error evaluation module is used to simulate the docking process of new energy vehicle components on the topological structure model of new energy vehicle components through the process flow structure model, so as to obtain the docking simulation data of new energy vehicle components; and to evaluate the docking accuracy error of components on the docking simulation data of new energy vehicle components, so as to obtain the docking component accuracy error data;

[0192] The process model optimization module is used to identify the component docking precision error process according to the docking component precision error data and the influencing factors of the sheet metal structure deformation, so as to obtain the precision error process data, and optimize the process parameters of the process flow structure model according to the precision error process data, so as to obtain the sheet metal process flow optimization model.

[0193] The process model generation system of the new energy sheet metal process of the present invention can realize the process model generation method of any new energy sheet metal process of the present invention, and is used to combine the operation and signal transmission medium between various modules to complete the process model generation method of the new energy sheet metal process. The internal modules of the system cooperate with each other to reduce the precision error of the sheet metal process.

[0194] Therefore, the embodiments should be regarded as illustrative and non-restrictive from all points, and the scope of the present invention is limited by the appended claims rather than the above description, and it is therefore intended that all changes falling within the meaning and range of equivalent elements of the application documents are included in the present invention.

[0195] The above description is only a specific embodiment of the present invention, so that those skilled in the art can understand or implement the present invention. 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 invention. Therefore, the present invention will not be limited to the embodiments shown herein, but should conform to the widest scope consistent with the principles and novel features invented herein.

Claims

1. A method for generating a process model for a new energy sheet metal process, characterized in that: The following steps are involved: Step S1: acquiring historical new energy vehicle body manufacturing data, and extracting vehicle body component structural features from the historical new energy vehicle body manufacturing data, thereby obtaining new energy vehicle body component structural data; performing new energy vehicle component topological structure analysis based on the new energy vehicle body component structural data, thereby obtaining a new energy vehicle component topological structure model; Step S2: extracting sheet metal process features from historical new energy vehicle body manufacturing data to obtain body manufacturing sheet metal process data, and performing sheet metal process flow structure modeling based on the body manufacturing sheet metal process data to obtain a process flow structure model; Step S3: estimating the sheet metal structure deformation amount of the process flow based on the historical new energy vehicle body manufacturing data based on the process flow structure model, thereby obtaining the sheet metal structure deformation amount data of the process flow; Calculate the sheet metal structure deformation variable influencing factors for the sheet metal process data of the body manufacturing and the sheet metal structure deformation variable data of the process flow, so as to obtain the sheet metal structure deformation variable influencing factors; Step S4: simulate the new energy vehicle component docking process on the new energy vehicle component topology structure model through the process flow structure model, so as to obtain the new energy vehicle component docking simulation data; evaluate the component docking accuracy error on the new energy vehicle component docking simulation data, so as to obtain the docking component accuracy error data. Step S4 is specifically as follows: Step S41: extracting the structure of the spliced ​​components from the topological structure model of the new energy vehicle components, thereby obtaining the spliced ​​structure model of the new energy vehicle components; Step S42: performing an automobile component docking process simulation on the new energy automobile component splicing structure model according to the process flow structure model, thereby obtaining new energy automobile component docking simulation data; Step S43: performing automobile component similarity calculation according to the new energy automobile component docking simulation data, thereby obtaining automobile component similarity data, and performing similar automobile component docking simulation clustering on the new energy automobile component docking simulation data according to the automobile component similarity data, thereby obtaining new energy automobile component docking simulation clustering data; Step S44: performing docking accuracy error calculation according to the new energy vehicle component docking simulation clustering data, thereby obtaining simulated docking accuracy error data; Step S45: evaluating the docking component accuracy error of the new energy vehicle component splicing structure model according to the simulated docking accuracy error data, thereby obtaining the docking component accuracy error data; Step S5: Perform component docking precision error process identification based on the docking component precision error data and the sheet metal structure deformation influencing factors to obtain precision error process data, and optimize process parameters of the process flow structure model based on the precision error process data to obtain a sheet metal process flow optimization model.

2. The process model generation method of the new energy sheet metal process according to claim 1 is characterized in that: Step S1 is specifically as follows: Step S11: acquiring historical new energy vehicle body manufacturing data, and extracting vehicle body component structural features from the historical new energy vehicle body manufacturing data, thereby obtaining new energy vehicle body component structural data; Step S12: extracting structural material features from the structural data of the new energy vehicle body parts to obtain body structure material data, and performing material property analysis based on the body structure material data to obtain body structure material property data; Step S13: classifying the body component structure data of the new energy vehicle according to the body component structure connection mode, thereby obtaining component structure welding data and component structure splicing data; Step S14: integrating the structural connection relationship of the vehicle body components according to the component structure welding data and the component structure splicing data, thereby obtaining the structural connection data of the vehicle body components; Step S15: Performing a topological structure analysis on the body parts based on the body part structure connection data to obtain a body part topological structure model, and integrating the body structure characteristics of the body part topological structure model according to the body structure material characteristic data to obtain a new energy vehicle component topological structure model.

3. The process model generation method of the new energy sheet metal process according to claim 2 is characterized in that: Step S13 is specifically as follows: Step S131: classifying the component structure data of the new energy vehicle body to obtain the connection component structure data and the main component structure data; Step S132: extracting the main component structure connection structure features from the main component structure data, thereby obtaining the main component connection structure data, and calculating the connection structure similarity between the component connection structure data and the main component connection structure data, thereby obtaining the connection structure similarity data; Step S133: integrating the component structure splicing relationship of the connection component structure data and the main component structure data according to the connection structure similarity data, thereby obtaining component structure splicing data; Step S134: performing component structure roughness calculation on the main component structure data to obtain component structure roughness data, and performing structure roughness threshold statistics based on the component structure roughness data to obtain a structure roughness threshold; Step S135: clustering the main component structure data into high-roughness structures according to the structure roughness threshold, thereby obtaining high-roughness structure data; Step S136: Integrate the high-roughness structure connection relationship of the main component structure data according to the high-roughness structure data, so as to obtain component structure welding data.

4. The process model generation method of the new energy sheet metal process according to claim 1 is characterized in that: Step S2 is specifically as follows: Step S21: performing body mode manufacturing process clustering on historical new energy vehicle body manufacturing data to obtain body mode manufacturing process data, and performing sheet metal process flow feature extraction and sheet metal process parameter feature extraction on the body mode manufacturing process data to obtain sheet metal process flow data and sheet metal process parameter data; Step S22: performing abnormal sheet metal process parameter identification according to the sheet metal process parameter data, thereby obtaining abnormal sheet metal process parameter data, and performing abnormal sheet metal process parameter correction on the sheet metal process parameter data according to the abnormal sheet metal process parameter data, thereby obtaining sheet metal process correction parameter data; Step S23: performing a body mode sheet metal process parameter trend analysis according to the sheet metal process correction parameter data, thereby obtaining body mode sheet metal process parameter trend data; Step S24: integrating the process parameters of the sheet metal process data according to the sheet metal process parameter trend data of the vehicle body model, thereby obtaining the sheet metal process data of the vehicle body manufacturing; Step S25: Perform sheet metal process flow structure modeling according to the body manufacturing sheet metal process data, so as to obtain a process flow structure model.

5. The process model generation method of the new energy sheet metal process according to claim 4 is characterized in that: Step S22 is specifically as follows: Step S221: performing frequency domain conversion on the sheet metal process parameter data to obtain a sheet metal process parameter spectrum; Step S222: integrating the frequency amplitude characteristics according to the sheet metal process parameter spectrum. Thus, the frequency amplitude data of sheet metal process is obtained; Step S223: performing frequency amplitude threshold statistics based on the sheet metal process frequency amplitude data, thereby obtaining the sheet metal process frequency amplitude threshold; Step S224: clustering the abnormal frequency amplitude spectrum of the sheet metal process parameter spectrum according to the sheet metal process frequency amplitude threshold, so as to obtain the sheet metal process abnormal frequency amplitude spectrum; Step S225: performing time domain conversion on the abnormal frequency amplitude spectrum of the sheet metal process, thereby obtaining abnormal sheet metal process parameter data; Step S226: performing abnormal sheet metal process parameter correction on the sheet metal process parameter data according to the abnormal sheet metal process parameter data, thereby obtaining sheet metal process correction parameter data.

6. The process model generation method of the new energy sheet metal process according to claim 1 is characterized in that: Step S3 is specifically as follows: Step S31: estimating the sheet metal structure deformation amount of the historical new energy vehicle body manufacturing data, thereby obtaining the sheet metal structure deformation amount data of the body manufacturing; Step S32: dividing the sheet metal structure deformation data of the vehicle body manufacturing into sheet metal structure deformation data of the process flow according to the process structure model, thereby obtaining the sheet metal structure deformation data of the process flow; Step S33: performing principal component analysis on the sheet metal process data of the vehicle body manufacturing, thereby obtaining principal component combination data of sheet metal process parameters; Step S34: constructing a linear regression mathematical model based on the sheet metal structural deformation variable data of the process flow and the principal component combination data of the sheet metal process parameters, thereby obtaining a structural deformation variable regression model; Step S35: calculating the sheet metal process parameter regression coefficients for the sheet metal structure deformation variable data of the process flow and the sheet metal process parameter principal component combination data according to the structure deformation variable regression model, thereby obtaining the sheet metal process parameter regression coefficient data; Step S36: Select parameters influencing the sheet metal structure deformation variable based on the sheet metal process parameter regression coefficient data, so as to obtain the sheet metal structure deformation variable influencing factors.

7. The process model generation method of the new energy sheet metal process according to claim 1 is characterized in that: Step S45 is specifically as follows: Step S451: extracting component accuracy error structural features according to the simulated docking accuracy error data, thereby obtaining component accuracy error structural data; Step S452: performing structural accuracy error statistics according to the component accuracy error structural data, thereby obtaining component high-accuracy error structural data and component low-accuracy error structural data; Step S453: classifying the high-precision error amount structure data of the component and the low-precision error amount structure data of the component according to the proportion of the high-precision error amount structure, so as to obtain the high-precision error amount structure high proportion component data and the low-precision error amount structure high proportion component data; Step S454: performing component structure association on the new energy vehicle component splicing structure model according to the high-precision error amount structure high-ratio component data, thereby obtaining high-precision error docking component data; performing component structure association on the new energy vehicle component splicing structure model according to the low-precision error amount structure high-ratio component data, thereby obtaining low-precision error docking component data; Step S455: merging the high-precision error docking component data and the low-precision error docking component data to obtain docking component precision error data.

8. The process model generation method of the new energy sheet metal process according to claim 1 is characterized in that: Step S5 is specifically as follows: Step S51: extracting high-precision error component structure features according to the docking component precision error data, thereby obtaining high-precision error component structure data; Step S52: extracting component structure process flow parameters from the high-precision error component structure data according to the process flow structure model, thereby obtaining error component structure process flow parameter data; Step S53: Associating the process error influencing parameters with the process error influencing parameters of the error component structure process parameter data according to the sheet metal structure deformation influencing factors, thereby obtaining the precision error process data; Step S54: optimizing the process flow structure model for minimum precision error of process parameters according to the precision error process data, thereby obtaining a sheet metal process flow optimization model.

9. A process model generation system for new energy sheet metal process, characterized in that: A process model generation method for a new energy sheet metal process according to claim 1, wherein the process model generation system for the new energy sheet metal process comprises: The component topology structure analysis module is used to obtain historical new energy vehicle body manufacturing data, and extract the structural features of the vehicle body components from the historical new energy vehicle body manufacturing data, so as to obtain the structural data of the new energy vehicle body components; perform the topology structure analysis of the new energy vehicle components based on the structural data of the new energy vehicle body components, so as to obtain the topology structure model of the new energy vehicle components; The sheet metal process flow structure modeling module is used to extract sheet metal process features from historical new energy vehicle body manufacturing data, thereby obtaining body manufacturing sheet metal process data, and to perform sheet metal process flow structure modeling based on the body manufacturing sheet metal process data, thereby obtaining a process flow structure model; The deformation variable influencing factor calculation module is used to estimate the deformation variable of the sheet metal structure of the process flow for the historical new energy vehicle body manufacturing data based on the process flow structure model, so as to obtain the sheet metal structure deformation variable data of the process flow; the sheet metal structure deformation variable influencing factor is calculated for the body manufacturing sheet metal process data and the sheet metal structure deformation variable data of the process flow, so as to obtain the sheet metal structure deformation variable influencing factor; The docking accuracy error evaluation module is used to simulate the docking process of new energy vehicle components on the topological structure model of new energy vehicle components through the process flow structure model, so as to obtain the docking simulation data of new energy vehicle components; and to evaluate the docking accuracy error of components on the docking simulation data of new energy vehicle components, so as to obtain the docking component accuracy error data; The process model optimization module is used to identify the component docking precision error process according to the docking component precision error data and the influencing factors of the sheet metal structure deformation, so as to obtain the precision error process data, and optimize the process parameters of the process flow structure model according to the precision error process data, so as to obtain the sheet metal process flow optimization model.

Images