A parameterized modeling method based on UGNX software
By filtering and modifying key feature parameters in UGNX software and using Python and NXOpen API to implement parametric modeling, the problem of low efficiency in product series design was solved, defect-free models were generated, and design efficiency and data quality were improved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- NANTONG UNIV
- Filing Date
- 2026-03-26
- Publication Date
- 2026-07-24
AI Technical Summary
In product serialization design and customized production, existing technologies require frequent model modifications and simulation verification, resulting in long design iteration cycles, low efficiency, and many invalid variations, leading to a serious bottleneck in mold design efficiency.
Based on UGNX software, a series of product models are automatically generated by filtering and modifying key feature parameters related to defects. Parametric modeling is achieved using Python interface and NXOpen API. Combined with finite element analysis and MATLAB post-processing scripts, defect-free models are quickly generated.
It significantly improves the efficiency of serialized batch design, provides a high-quality model data foundation, supports subsequent CAE analysis and intelligent design assistance, and shortens the design cycle.
Smart Images
Figure CN122452208A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of computer-aided engineering design technology, and in particular to a parametric modeling method based on UGNX software. Background Technology
[0002] In product development and engineering design, computer-aided design software has become an indispensable core technology due to its powerful 3D modeling tools. While computer-aided software greatly improves design accuracy and visualization capabilities, enabling engineers to build complex product models, design work often faces numerous challenges in practical engineering applications. During product serialization design, customized production, or optimization iterations, it is often necessary to frequently modify the same basic model based on different design parameters, dimensional specifications, or functional requirements.
[0003] When a product series includes multiple specifications or needs to meet highly customized requirements, designers often need to repeatedly perform cyclical operations of parameter modification, model updates, and simulation verification, a process that is tedious and prone to errors. Common parametric modeling often generates all possible parameter combinations indiscriminately, leading to a large number of invalid variants with production defects. These invalid variants still need to be eliminated one by one through subsequent simulation analysis or actual production trials. This results in a serious efficiency bottleneck in key aspects such as mold design during product series development. Design iteration cycles are long, and it is difficult to quickly develop mold solutions covering the entire product range. Summary of the Invention
[0004] Purpose of the invention: In order to overcome the shortcomings of the existing technology, the present invention provides a parametric modeling method based on UGNX software, which can automatically generate all possible qualified products in a series by filtering and modifying key feature parameters related to defects on a basic model, thereby significantly improving the efficiency of serialized batch design and the efficiency of finding the optimal feature combination.
[0005] Technical solution: To achieve the above objectives, the present invention provides a parametric modeling method based on UGNX software, comprising the following steps:
[0006] S1. Create a basic model of the forging, and combine all the features that make up the basic model into a total feature set.
[0007] S2. Based on the preset defect formation mechanism, identify the defect-related features in the total feature set, determine the importance of the impact of each defect-related feature on the existing defects of the basic model after adjustment, select key features from the defect-related features, and form a follower feature set with the remaining defect-related features.
[0008] S3. In the set of passive features, find passive features with associated dimensions for each key feature, and construct an adjustable subset of key features and corresponding passive features.
[0009] S4. Based on the constraint relationships between the features of each adjustable subset, and in combination with practical engineering experience and design specifications, determine the actual value range of each key feature.
[0010] S5. Based on the actual value range, write and execute a parametric modeling script. Randomly extract size parameters from the corresponding actual value range for each key feature. On the basis of the basic model, modify the size parameters of each key feature to the extracted size parameters, and synchronously modify the driven features related to the key features based on the associated size relationship. After modification, save as a new model.
[0011] S6. Repeat step S5 to generate a variety of new models with different size combinations on a large scale.
[0012] Furthermore, in step S2, the screening of the key features includes the following steps:
[0013] S21. Based on extensive practical production experience, roughly summarize the types and distribution areas of defects in the forgings, and set these defects as preset defects;
[0014] S22. Based on the type of preset defect, adopt the corresponding prediction method to quickly locate the defect location and determine its severity. Based on the predicted defect location and severity, analyze the main physical causes of the preset defect and list the features involved in the physical causes as defect-related features.
[0015] S23. Among the multiple defect-related features, by analyzing the physical behaviors of the material in different feature regions during the simulation process that are related to causing defects, the importance of each defect-related feature in optimizing the preset defect is determined, and the defect-related features that have a significant impact on the preset defect are listed as key features.
[0016] Further, in step S22, when the preset defect is a material filling deficiency defect, the forging process simulation is performed on the basic model to locate all defect locations and corresponding severity of the basic model; based on the defect locations and severity obtained from the simulation results, a discretized network analysis is further performed to analyze and obtain the closely related characteristics of multiple physical factors of the defect location as the defect-related characteristics of the material filling deficiency defect.
[0017] Furthermore, after establishing numerous defect-related characteristics of the aforementioned material insufficiency defects, the key characteristics corresponding to the material insufficiency defects are screened out by analyzing the material flow trajectory, stress distribution, and deformation in different characteristic regions during the simulation process.
[0018] Further, in step S22, when the preset defect is a crack defect, the cumulative damage value D of different regions on the basic model is calculated by the normalized damage model, the region with a D value greater than a preset threshold is located as the defect location, and the severity is determined according to the size of the D value; based on the calculated defect location and severity, a discretized network analysis is further performed, and the analysis yields the closely related features of multiple physical factors of the defect location as the defect-related features of the material insufficiency defect.
[0019] Furthermore, after establishing the defect-related characteristics of numerous crack defects, the stress distribution and deformation of different characteristic regions are analyzed to eliminate crack defects caused by non-characteristic factors and screen out the key characteristics corresponding to the crack defects.
[0020] Furthermore, in step S4, the theoretical value range of the key feature is first determined under the condition that it does not cause feature interference or model error. Then, the theoretical value range is narrowed by combining the technical requirements of the key feature itself, industry standards, and its impact on defect formation, so as to obtain the actual value range of the key feature.
[0021] Furthermore, in step S5, a parametric modeling script is written based on the Python interface provided by UG, and the Siemens NX parts are automatically modified by parameter-driven modification through NXOpenAPI. After the feature parameter modification module is written, the execution directory, model import, model export, model closing and batch generation functions are added to the parametric modeling script to expand the script content and thus realize parametric modeling.
[0022] Furthermore, in step S5, the parametric modeling script is executed through the UG visualization interface. This script uses Python to call NXOpenAPI in a loop, randomly extracts dimensional parameters from a preset range, modifies the feature parameters in the template file using a rollback editing mechanism, and after ensuring geometric updates and system cleanup, simultaneously saves the modified scheme as an independent prt file, exports high-precision upper and lower mold STL files, and finally closes the modified file through a memory reclamation mechanism, thus realizing a large-scale, silent parametric modeling process.
[0023] Furthermore, after implementing the parametric modeling, the selected key features are verified using the built-in batch simulation function of the finite element software. After obtaining the simulation results, a MATLAB-based post-processing script is used to identify defects in the forging to verify the correctness of the key features. The parametric modeling script is then iteratively updated based on the verification results.
[0024] Beneficial Effects: This invention provides a parametric modeling method based on UGNX software. Using UGNX as the system foundation, engineers can directly utilize the software's 3D modeling capabilities to create basic models, which can then be directly used as input for batch design, offering convenience and speed. By selecting and modifying key feature parameters related to defects on the input basic model, a series of structured, parametric model files can be generated in batches and efficiently, significantly improving the efficiency of serialized batch design and the efficiency of finding optimal feature combinations. These high-quality models provide a solid data foundation for subsequent CAE analysis, design optimization, and even the application of cutting-edge technologies such as intelligent design assistance combined with machine learning, demonstrating broad prospects for expanded applications. Attached Figure Description
[0025] Figure 1 This is a flowchart of the parametric modeling method for product series design based on UGNX software, as described in this invention.
[0026] Figure 2 This is a structural diagram of a spacer bar clamp model according to an embodiment of the present invention.
[0027] Figure 3 This is a visualization result of incomplete filling defects based on a MATLAB post-processing script in one embodiment of the present invention.
[0028] Figure 4 This is a visualization result of crack defects calculated based on a MATLAB post-processing script and a normalized damage model in one embodiment of the present invention.
[0029] Figure 5 The relevant screening features of the spacer bar clamp are shown in one embodiment of the present invention.
[0030] Figure 6 This is a portion of the feature modification module in the parameterization script for the spacer bar clamp according to an embodiment of the present invention.
[0031] Figure 7 This is a functional module part of the parameterization script for the spacer bar clamp in one embodiment of the present invention.
[0032] Figure 8 This is a structural diagram of the initial rounded corner feature of the cross-axis region of the spacer bar clamp model according to an embodiment of the present invention.
[0033] Figure 9 The diagram shows the structure of the fillet feature of the cross axis region of the spacer bar clamp model according to an embodiment of the present invention when the actual value range is taken as two end values. (a) corresponds to R=1 and (b) corresponds to R=7. Detailed Implementation
[0034] The invention will now be further described with reference to the accompanying drawings.
[0035] As attached Figure 1-8 The parametric modeling method based on UGNX software, taking the spacer bar clamp forging for ultra-high voltage power transmission as an example, uses dynamic combination and adjustment of parameters based on key feature extraction to create a batch of defect-free serialized models quickly and efficiently. This model is used to build a forging die design candidate library, facilitating adaptive optimization design for the same series of products in different application scenarios. Specifically, it includes the following steps:
[0036] S1. Create a basic model of the spacer bar clamp forging, and assemble all features constituting this basic model into a total feature set. The spacer bar clamp is used in power transmission to fix and connect sub-conductors, preventing them from galloping and thus reducing the corona effect caused by galloping. The corona effect not only affects power transmission efficiency but can also lead to discharge phenomena and equipment damage. Therefore, the spacer bar clamp needs to possess high mechanical and fatigue performance.
[0037] S2. Based on extensive practical production experience, it was found that due to deficiencies in the mold structure design, forging defects such as insufficient filling and micro-cracks existed in the spacer bar clamps during actual production. Therefore, these defects were set as preset defects. Furthermore, as... Figure 3 As shown, the material insufficiency defect is mainly concentrated in the convex circular region 11 of the spacer bar clamp, while the crack defect is mainly concentrated in the cross axis region 12. Based on the formation mechanism of the preset defect, the defect-related features in the total feature set are identified. By judging the importance of the impact of each defect-related feature adjustment on the existing defects of the basic model, key features are selected from the defect-related features, and the remaining defect-related features are used to form a follower feature set.
[0038] In this embodiment, in order to quickly locate the location and severity of defects in forgings, a corresponding prediction method is triggered according to the type of preset defects to locate the defect location and determine the severity. Based on the predicted defect location and severity, the main physical causes of the preset defects are analyzed, and the features involved in the physical causes are listed as defect-related features.
[0039] To address the incomplete filling defect, a finite element simulation is first triggered to simulate the forging process of the basic model of the spacer bar clamp. A MATLAB-based post-processing script is then used to analyze the simulation data, quickly obtaining the locations and severity of all defects in the forging. The corresponding forging defects can be obtained by inputting a simulation file. The incomplete filling defect and its corresponding severity in the spacer bar clamp are shown below. Figure 3 As shown. Further discretized mesh analysis revealed that the incomplete filling phenomenon is closely related to the bridge height and bin width characteristics at the convex circle position. Therefore, these two features were automatically identified as defect-related features.
[0040] After establishing the defect-related features, the importance of each feature in forging defect optimization was determined by further analyzing the flow trajectory, stress distribution, and deformation of the material in different feature regions during the simulation. The analysis results show that the hopper feature is mainly used to achieve material recovery and material blocking functions; the existing feature size is sufficient to meet this requirement, therefore no major modifications are needed. In contrast, the bridge feature size has a significant impact on forging defects. During forging, the existing bridge height causes material to flow out of the cavity prematurely, leading to material loss in the convex area during subsequent forging, resulting in incomplete filling. Considering these factors, bridge feature 10 was identified as a key feature.
[0041] For crack defects, the cumulative damage value D is first calculated based on a normalized damage model to characterize the degree of material damage. The closer the damage value is to 1, the more severe the material damage, approaching fracture. The calculation of D is as follows:
[0042]
[0043] In the formula: This is the maximum yield stress of the material, i.e., the material's ultimate strength. Equivalent stress; Plastic strain is the permanent, unrecoverable deformation of a material caused by external forces.
[0044] Calculations showed that the D values of both the cross-axis region 12 and the rib region 13 were greater than 0.6, indicating the presence of microcracks. Figure 4As shown, the red circles mark cracks in the cross-axis region, and the green circles mark cracks in the rib region. Further discretization mesh analysis revealed that the microcracks in the cross-axis region are caused by significant stress concentration and are closely related to improper fillet feature dimensions in this region. Therefore, the fillet feature in this region is classified as a defect-related feature. The microcracks in the rib region are caused by stress concentration due to the arc feature and are also classified as defect-related defects. Further analysis of stress distribution and deformation revealed that the stress distribution in the rib region is uniform, with no stress concentration. Therefore, the cracks in this region are mainly related to the forging impact force of the equipment during production and are not strongly correlated with the fillet feature. However, significant stress concentration exists in the cross-axis region, and this phenomenon is closely related to fillet feature 20. Therefore, fillet feature 20 in the cross-axis region is classified as a key feature.
[0045] S3. In the set of driven features, find driven features with associated dimensions for each key feature, and construct an adjustable subset of the key features and their corresponding driven features. Specifically, based on the technical drawing requirements, for example, considering the proportional coordination of local surfaces, if multiple features within a surface area have a fixed proportional or difference in associated dimensions, then the key features and driven features within that surface area can be constructed into an adjustable subset. If two or more key features have associated dimensions with the same other related feature, then these two or more key dimensions are all constructed into an adjustable subset. If a non-key related feature does not have an associated dimension relationship with any key feature, then this related feature will be constructed as a separate adjustable subset.
[0046] S4. Based on the constraint relationship between the features of each adjustable subset, combined with actual engineering experience and design specifications, determine the actual value range of each key feature; first, determine the theoretical value range of the key feature under the condition that it does not cause feature interference and model error, and then, combined with the technical requirements of the key feature itself, industry standards and its impact on defect formation analysis, narrow the theoretical value range to obtain the actual value range of the key feature.
[0047] Selected features such as Figure 5As shown, for the fillet feature in the cross-axis region, the theoretical value space of its characteristic parameters is the interval [1.0, 10.0], provided that it does not cause feature interference or model errors. After determining the theoretical value space of the characteristic parameters, the value space of the characteristic parameters is further determined by combining the relevant technical requirements of the spacer bar clamp drawings and industry standards. For this fillet feature, since the cross-axis region is mainly used to fix and connect wires, according to industry standards, its maximum fillet radius is 7.0. However, to avoid generating larger crack defects, the minimum fillet radius used in production is 1.0. Therefore, the actual value space of the characteristic parameters of the cross-axis fillet feature is the interval [1.0, 7.0]. As for the bridge height feature, its theoretical value space is [0.0, 10.0] without causing feature interference or model errors. However, according to the technical specifications, the bridge height should be within the range of [2.0, 5.0]. Considering that a bridge height that is too low will reduce the service life of the mold, while a bridge height that is too high will cause material to be lost too early, making it difficult to fill the cavity and thus exacerbating the incomplete filling phenomenon, its actual value space is set to [2.5, 3.5].
[0048] S5. Based on the actual value range, write and execute a parametric modeling script. Randomly extract size parameters from the corresponding actual value range for each key feature. On the basis of the basic model, modify the size parameters of each key feature to the extracted size parameters, and synchronously modify the driven features related to the key features based on the associated size relationship. After modification, save as a new model.
[0049] Specifically, parametric modeling scripts are written based on the Python interface provided by UG, and automated parameter-driven modification of Siemens NX parts is achieved through the NXOpen API. First, the model file is opened according to the specified path and the modeling module is switched. Then, the rollback editing mechanism is used to accurately locate specific features, replace their values with the input parameters, and automatically clean up temporary objects and expressions generated during the execution process after completing the geometry update and feature submission, so as to ensure the stability of the model after parameter adjustment.
[0050] like Figure 5 As shown, after completing the feature parameter modification module, the parametric modeling script is expanded with functions for execution directory, model import, model export, model closure, and batch generation. This expands the script content and enables parametric modeling. The parametric script for the spacer bar clamp is shown below. Figure 7 As shown.
[0051] After the system filters out key features and determines their corresponding actual value ranges, it stores them in a folder for script calls. Then, the parametric modeling script is triggered to run. The script uses Python to call NXOpenAPI in a loop, randomly extracts dimensional parameters from the corresponding actual value range, modifies the key feature parameters and driven feature parameters in the template file using a rollback editing mechanism, and after ensuring geometric updates and system cleanup, saves the modified scheme as an independent prt file and exports high-precision upper and lower mold STL files. Finally, the modified files are closed through a memory reclamation mechanism, thus realizing a large-scale, silent parametric modeling process.
[0052] After implementing the parametric modeling, the selected key features are verified using the built-in batch simulation function of the finite element software. After obtaining the simulation results, a MATLAB-based post-processing script is used to identify defects in the forging to verify the correctness of the key features. The parametric modeling script is then iteratively updated based on the verification results.
[0053] S6. Repeat step S5 to generate a variety of new models with different size combinations on a large scale. These new models are all preferred structures for forgings without preset defects, and together they form a selection database to assist in the design of forging dies.
[0054] This solution constructs a parametric modeling system to assist in forging die design. The system takes a basic forging model, common production defect types, technical drawing requirements, and industry standards as input. Based on the input basic model, it automatically generates and outputs a series of all possible qualified products by filtering and modifying key feature parameters related to defects. These high-quality output models provide a solid data foundation for subsequent CAE analysis, design optimization, and even the application of cutting-edge technologies such as intelligent design assistance combined with machine learning, demonstrating broad prospects for expanded applications.
[0055] The above description is only a preferred embodiment of the present invention. It should be noted that for those skilled in the art, several improvements and modifications can be made without departing from the above principles of the present invention, and these improvements and modifications are also considered to be within the scope of protection of the present invention.
Claims
1. A parametric modeling method based on UGNX software, characterized in that, Includes the following steps: S1. Create a basic model of the forging and combine all the features that constitute the basic model into a total feature set; S2. Based on the preset defect formation mechanism, identify the defect-related features in the total feature set, determine the importance of the impact of each defect-related feature on the existing defects of the basic model after adjustment, select key features from the defect-related features, and form a follower feature set with the remaining defect-related features. S3. In the set of passive features, find passive features with associated dimensions for each key feature, and construct an adjustable subset of key features and corresponding passive features; S4. Based on the constraint relationship between the features of each adjustable subset, and combined with practical engineering experience and design specifications, determine the actual value range of each key feature. S5. Based on the actual value range, write and execute a parametric modeling script. By randomly extracting size parameters from the corresponding actual value range for each key feature, modify the size parameters of each key feature to the extracted size parameters on the basis of the basic model, and synchronously modify the driven features related to the key features based on the associated size relationship. Save the modified file as a new model; S6. Repeat step S5 to generate a variety of new models with different size combinations on a large scale.
2. The parametric modeling method based on UGNX software according to claim 1, characterized in that: In step S2, the screening of the key features includes the following steps: S21. Based on extensive practical production experience, roughly summarize the types and distribution areas of defects in the forgings, and set these defects as preset defects; S22. Based on the type of preset defect, adopt the corresponding prediction method to quickly locate the defect location and determine its severity. Based on the predicted defect location and severity, analyze the main physical causes of the preset defect and list the features involved in the physical causes as defect-related features. S23. Among the multiple defect-related features, by analyzing the physical behaviors of the material in different feature regions during the simulation process that are related to causing defects, the importance of each defect-related feature in optimizing the preset defect is determined, and the defect-related features that have a significant impact on the preset defect are listed as key features.
3. The parametric modeling method based on UGNX software according to claim 2, characterized in that: In step S22, when the preset defect is a material filling deficiency defect, the forging process simulation is performed on the basic model to locate all defect locations and their corresponding severity in the basic model; based on the defect locations and severity obtained from the simulation results, a discretized network analysis is further performed to analyze and obtain the closely related features of multiple physical factors of the defect location as the defect-related features of the material filling deficiency defect.
4. The parametric modeling method based on UGNX software according to claim 3, characterized in that: After establishing numerous defect-related characteristics of the aforementioned material insufficiency defects, the key characteristics corresponding to the material insufficiency defects are screened out by analyzing the material flow trajectory, stress distribution, and deformation in different characteristic regions during the simulation process.
5. The parametric modeling method based on UGNX software according to claim 2, characterized in that: In step S22, when the preset defect is a crack defect, the cumulative damage value D of different regions on the basic model is calculated by the normalized damage model. Regions with D values greater than a preset threshold are located as defect locations, and the severity is determined based on the magnitude of the D value. Based on the calculated defect location and severity, a discretized network analysis is further performed to analyze and obtain the closely related characteristics of multiple physical factors of the defect location as the defect-related characteristics of the material insufficiency defect.
6. The parametric modeling method based on UGNX software according to claim 5, characterized in that: After establishing the defect-related characteristics of numerous crack defects, the stress distribution and deformation of different characteristic regions are analyzed to eliminate crack defects caused by non-characteristic factors and screen out the key characteristics corresponding to the crack defects.
7. The parametric modeling method based on UGNX software according to claim 1, characterized in that: In step S4, the theoretical value range of the key feature is first determined under the condition that it does not cause feature interference or model error. Then, the theoretical value range is narrowed by combining the technical requirements of the key feature itself, industry standards, and its impact on defect formation, so as to obtain the actual value range of the key feature.
8. The parametric modeling method based on UGNX software according to claim 1, characterized in that: In step S5, a parametric modeling script is written based on the Python interface provided by UG, and the Siemens NX parts are automatically modified by parameter-driven modification through the NXOpen API. After the feature parameter modification module is written, the execution directory, model import, model export, model closing and batch generation functions are added to the parametric modeling script to expand the script content and thus realize parametric modeling.
9. A parametric modeling method based on UGNX software according to claim 8, characterized in that: In step S5, the parametric modeling script is executed through the UG visualization interface. The script uses Python to call NXOpenAPI in a loop, randomly extracts dimensional parameters from a preset range, modifies the feature parameters in the template file using a rollback editing mechanism, and after ensuring geometric updates and system cleanup, saves the modified scheme as an independent prt file and exports high-precision upper and lower mold STL files. Finally, the modified file is closed through a memory reclamation mechanism, thus realizing a large-scale, silent parametric modeling process.
10. A parametric modeling method based on UGNX software according to claim 9, characterized in that: After implementing the parametric modeling, the selected key features are verified using the built-in batch simulation function of the finite element software. After obtaining the simulation results, a MATLAB-based post-processing script is used to identify defects in the forging to verify the correctness of the key features. The parametric modeling script is then iteratively updated based on the verification results.