Fixture three-dimensional parametric modeling method for welding fixture design

Abstract

The invention relates to the technical field of three-dimensional modeling, in particular to a fixture three-dimensional parametric modeling method for welding fixture design, which comprises the following steps of: synthesizing the welding deformation probability of all grid nodes in each cluster and the proportion of all grid nodes in each cluster in all grid nodes in a three-dimensional grid model to obtain the welding deformation probability of all grid nodes in the three-dimensional grid model; the risk degree is evaluated; a collision detection algorithm is adopted to evaluate the collision degree under each welding fixture layout scheme; a path planning algorithm is adopted to determine the mobility rate under each welding fixture layout scheme; and based on the collision degree and the mobility rate and in combination with the risk degree, determining a welding robust score under each welding fixture layout scheme so as to obtain a three-dimensional model of the optimal layout of the welding fixture. The problem that traditional clamp design depends on experience and repeated trial and error is solved, and the reliability of welding clamp design is improved.

Description

Technical Field

[0001] This application relates to the field of 3D modeling technology, specifically to a 3D parametric modeling method for welding fixture design. Background Technology

[0002] To meet the stringent requirements of modern manufacturing for the flexibility, efficiency, and consistent quality of welded product production, welding fixture design has evolved from two-dimensional static drawings to three-dimensional parametric intelligent modeling. By defining the size, positioning, and clamping relationship of the fixture as variable parameters, modern design methods have enabled design logic to drive the process. This not only greatly shortens the R&D cycle but also fundamentally ensures design quality and reusability, serving as a core technology for improving automated welding levels and production flexibility.

[0003] The core contradiction in welding fixture layout design lies in the trade-off between suppressing welding deformation and ensuring process feasibility. On the one hand, simplifying the fixture structure to reduce costs may reduce stress, but it may also lead to the ineffective release of welding thermal stress, thus exacerbating workpiece deformation. On the other hand, adding clamping points to control deformation may improve the constraint effect, but it will severely compress the operating space, causing tool movement interference and collisions. However, most current mainstream 3D modeling is limited to the static presentation of geometric representation and assembly relationships, failing to deeply integrate welding process data. This results in defects in the process feasibility of the design scheme, forcing the design process into a cycle of repeated trial and error and manual adjustments, reducing the reliability of welding fixture design. Summary of the Invention

[0004] To address the aforementioned technical problems, this application provides a three-dimensional parametric modeling method for welding fixture design, thereby resolving the existing issues.

[0005] The present application provides a three-dimensional parametric modeling method for welding fixture design, which adopts the following technical solution: One embodiment of this application provides a three-dimensional parametric modeling method for welding fixture design, the method comprising the following steps: Obtain the 3D model of each welding fixture layout scheme and perform mesh processing on it; Analyze the historical welding deformation probability of each grid node to predict the current welding deformation probability of each grid node; cluster all grid nodes, combine the current welding deformation probability of all grid nodes in each cluster, and the proportion of all grid nodes in each cluster in all grid nodes in the 3D mesh model to evaluate the risk level under each welding fixture layout scheme. A three-dimensional ligand model and a three-dimensional surface model are obtained for each welding fixture layout scheme. A collision detection algorithm is then used to evaluate the collision degree for each welding fixture layout scheme, based on the three-dimensional models. Path planning data and the three-dimensional spatial coordinates of the welding fixture are obtained for each welding fixture layout scheme. A path planning algorithm is then used to determine the mobility rate for each welding fixture layout scheme. Based on the collision degree and the mobility rate, the movement characteristic value for each welding fixture layout scheme is determined. Combined with the risk level, a welding robustness score for each welding fixture layout scheme is then determined. Based on the welding robustness score, a 3D model of the optimal layout of the welding fixture is obtained.

[0006] Preferably, the method for determining the current welding deformation probability of each grid node is as follows: The random forest algorithm takes all mesh nodes and their historical welding deformation probabilities in the 3D model under each welding fixture layout scheme as input and outputs the current welding deformation probability of each mesh node in the 3D model under each welding fixture layout scheme.

[0007] Preferably, the method for determining the distance metric during clustering is as follows: The current 3D coordinates of each grid node are multiplied by a preset first coefficient and the current welding deformation probability is multiplied by a preset second coefficient to form a quadruple. The distance is measured as the Euclidean distance between the grid node quadruples, where the preset second coefficient is greater than the preset first coefficient.

[0008] Preferably, the assessment of the risk level under each welding fixture layout scheme includes: In the 3D model of each welding fixture layout scheme, the comprehensive deformation risk value of each cluster is determined based on the welding deformation probability of all grid nodes in each cluster. The risk level of each welding fixture layout scheme is positively correlated with the comprehensive deformation risk value of each cluster and the proportion of all mesh nodes in each cluster in the three-dimensional mesh model.

[0009] Preferably, the comprehensive deformation risk value of each cluster is the average of the welding deformation probabilities of all grid nodes in each cluster.

[0010] Preferably, the collision degree under each welding fixture layout scheme is obtained by using a collision detection algorithm on the three-dimensional model, three-dimensional ligand model and three-dimensional surface model under each welding fixture layout scheme to obtain the number of collisions of the welding fixture under each welding fixture layout scheme.

[0011] Preferably, the method for determining the mobility rate under each welding fixture layout scheme is as follows: The path planning data and the three-dimensional spatial coordinates of the welding fixture under each welding fixture layout scheme are used as input to the path planning algorithm, and the optimal path is output. Based on the path planning data for each welding fixture layout scheme, the longest path length is obtained, and the result of dividing the optimal path length by the longest path length is used as the mobility rate for each welding fixture layout scheme.

[0012] Preferably, the movement characteristic value under each welding fixture layout scheme is negatively correlated with the collision degree and mobility rate under each welding fixture layout scheme.

[0013] Preferably, the welding robustness score under each welding fixture layout scheme is positively correlated with the movement characteristic value under each welding fixture layout scheme and negatively correlated with the risk level.

[0014] Preferably, the three-dimensional modeling for obtaining the optimal layout of the welding fixture includes: The three-dimensional models under a preset number of welding fixture layout schemes are used as input to the genetic algorithm. The welding robustness score under each welding fixture layout scheme is used as the fitness function value in the genetic algorithm, and the three-dimensional model under the optimal welding fixture layout is output.

[0015] This application has at least the following beneficial effects: This application first uses the random forest algorithm to predict the deformation probability of each grid node, then uses DBSCAN clustering to group discrete risk points into spatially significant risk clusters. Combining the risk intensity and spatial proportion of each cluster, the risk level of each welding fixture layout scheme is quantified, thus transforming the fuzzy deformation problem into a precise and quantifiable evaluation index, providing a scientific basis for fixture layout decisions. Furthermore, this application uses A... The path planning algorithm plans the optimal path to quantify the mobility rate, and the collision detection algorithm counts the number of collisions. The two are then integrated into a comprehensive movement feature value to characterize the process feasibility. Furthermore, the movement feature value is combined with the risk level to construct a welding robustness score, thereby realizing a global quantitative evaluation of the welding fixture layout scheme in terms of both quality reliability and process feasibility. Finally, this application uses the welding robustness score as the fitness function of the genetic algorithm. Through iterative optimization of the fixture position, it ultimately drives the 3D parametric modeling system to automatically generate the optimal fixture layout model with the lowest risk and highest accessibility, realizing an intelligent closed loop from evaluation to design and improving the reliability of welding fixture design. Attached Figure Description

[0016] To more clearly illustrate the technical solutions and advantages in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0017] Figure 1 A flowchart illustrating the steps of a three-dimensional parametric modeling method for welding fixture design, provided in one embodiment of this application; Figure 2 This is a schematic diagram of the welding robustness score extraction process provided in one embodiment of this application. Detailed Implementation

[0018] To further illustrate the technical means and effects adopted by this application to achieve the intended purpose of the invention, the following, in conjunction with the accompanying drawings and preferred embodiments, details the specific implementation, structure, features, and effects of a three-dimensional parametric modeling method for welding fixture design proposed in this application. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics in one or more embodiments can be combined in any suitable form.

[0019] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application pertains.

[0020] The following description, in conjunction with the accompanying drawings, details a specific scheme for a three-dimensional parametric modeling method for welding fixture design provided in this application.

[0021] One embodiment of this application provides a three-dimensional parametric modeling method for welding fixture design. Specifically, it provides the following method for three-dimensional parametric modeling of welding fixtures. Please refer to [link to relevant documentation]. Figure 1 The method includes the following steps: Step S1: Obtain the 3D model of each welding fixture layout scheme and mesh it.

[0022] To construct a high-fidelity, computable collaborative digital twin environment for welding fixtures, the core foundation lies in the systematic collection and fusion of multi-source heterogeneous data. This allows for the precise mapping of all elements of the physical world into a virtual space. This process begins with the digital construction of the geometric ontology: First, 3D models of each welding fixture layout scheme for machining workpieces of the same batch and specifications need to be exported from mainstream 3D CAD software (such as SolidWorks, CATIA, NX, etc.). In this embodiment, SolidWorks is used. The 3D model not only defines the macroscopic geometric shape of each component of the fixture but also includes the precise assembly relationships and constraints between them. Furthermore, the continuous solid model is discretized into a 3D mesh model composed of a massive number of tiny units through meshing.

[0023] The process of obtaining the 3D model and 3D mesh for each welding fixture layout scheme using CAD software is a well-known technology and will not be described in detail here.

[0024] Step S2: Analyze the historical welding deformation probability of each grid node to predict the current welding deformation probability of each grid node; cluster all grid nodes, combine the current welding deformation probability of all grid nodes in each cluster, and the proportion of all grid nodes in each cluster in all grid nodes in the 3D mesh model, to evaluate the risk level under each welding fixture layout scheme.

[0025] In the process of 3D modeling of welding fixtures, there are drawbacks such as uneven welding stress distribution and highly complex geometric features. Traditional modeling methods cannot effectively determine the distribution of welding deformation risks, resulting in a lack of specificity in fixture layout and easily leading to insufficient or over-constraint of some important parts. To solve this problem, this embodiment analyzes the historical welding deformation probability of each mesh node to predict the current welding deformation probability of each mesh node; it clusters all mesh nodes and combines the current welding deformation probability of all mesh nodes in each cluster with the proportion of all mesh nodes in each cluster in the 3D mesh model to evaluate the risk level of each welding fixture layout scheme. The specific process is as follows: First, the historical welding deformation probability of each grid node is obtained to predict the current welding deformation probability of each grid node. Specifically: In this embodiment, for workpieces of the same specifications in the same batch, multiple different welding fixture layout schemes are created. Using finite element analysis software, the welding process under different welding fixture layout schemes is simulated by thermo-mechanical coupling. After the simulation is completed, the finite element analysis software will calculate the welding deformation probability at each mesh node on the three-dimensional model during the virtual welding process. Therefore, the historical welding deformation probability of each mesh node in the three-dimensional model under each welding fixture layout scheme is obtained by using finite element simulation software to quantify the deformation risk of the welding fixture at each mesh node.

[0026] The process of obtaining the historical welding deformation probability using finite element simulation software is a well-known technique, and the specific acquisition process will not be described in detail here.

[0027] Furthermore, all grid nodes in the 3D model under each welding fixture layout scheme and their respective historical welding deformation probabilities are used as input to the random forest algorithm. In this embodiment, the number of decision trees is set to 100, the maximum depth is 15, and the minimum number of samples for internal node subdivision is 5. The random forest algorithm is trained and finally outputs the current welding deformation probability of each grid node in the 3D model under each welding fixture layout scheme.

[0028] It should be noted that the number of decision trees is set to 100 because the model's prediction accuracy tends to converge around this number. Increasing the number of trees further would lead to disproportionate computational overhead with minimal benefits. The maximum depth is set to 15 because too small a depth may prevent the model from fully learning the complex linear patterns of welding deformation, while too large a depth can easily lead to overfitting to the noise in the training data. A maximum depth of 15 provides a compromise between fitting complex patterns and having good generalization ability. The minimum number of samples for internal node subdivision is set to 5 instead of less to avoid the model generating overly fragmented decision rules for individual noisy data points, thereby enhancing the model's robustness, while also preventing the important local high-risk features from being ignored due to an excessively large number of samples.

[0029] The specific process of training the random forest algorithm is a well-known technique and will not be elaborated further.

[0030] Furthermore, this embodiment evaluates the risk level of each welding fixture layout scheme by clustering all mesh nodes, combining the welding deformation probability of all mesh nodes in each cluster, and the proportion of all mesh nodes in each cluster among all mesh nodes in the 3D mesh model. Specifically: All mesh nodes in the 3D model under each welding fixture layout scheme are used as input to the clustering algorithm. The 3D coordinates of each mesh node are multiplied by a preset first coefficient and the current welding deformation probability is multiplied by a preset second coefficient to form a quadruple. The distance is measured by the Euclidean distance between the mesh node quadruples. The preset second coefficient is greater than the preset first coefficient. The number of clusters is determined by the elbow rule. The neighborhood radius is set to a preset first value. The minimum number of points is set to a preset second value. Finally, all clusters are output.

[0031] It should be noted that there are many commonly used clustering algorithms. In this embodiment, the DBSCAN density clustering algorithm is used to cluster the grid nodes. In practical applications, as other implementation methods, implementers may also use other clustering methods such as the k-means clustering algorithm to cluster the grid nodes according to specific circumstances. This embodiment does not impose any special restrictions on the selection of clustering algorithms.

[0032] It should be noted that in this embodiment, the preset first coefficient is set to 1, and the preset second coefficient is set to 5. In this embodiment, if the preset second coefficient is too small, the weight of the welding deformation probability will not be increased enough, and it may still be unable to effectively suppress the dominant role of spatial distance. For a point with high risk but a slightly off-center location, the algorithm may still incorrectly classify it into a low-risk cluster because it is spatially closer to the low-risk point group, resulting in an insignificant clustering effect and unclear risk area division. If the preset value is too large, it will lead to overcorrection, and the clustering algorithm will forcibly pull all high-risk points together, regardless of how far apart they are in space, ultimately forming a huge and loose high-risk cluster. In this embodiment, the preset first value, i.e., the neighborhood radius, is 0.1, which is obtained from experience. The preset value of 5 is a well-tuned golden pair with the neighborhood radius of 0.1 mm. Therefore, in this embodiment, the preset value is set to 5.

[0033] Furthermore, it should be noted that in this embodiment, the second preset value, namely the minimum number of points, is set to 8. This is because the value of the neighborhood radius and the value of the minimum number of points work together. In this embodiment, the value of the neighborhood radius is 0.1 mm. Through experimental optimization, it was found that when the minimum number of points is 8, it can work together with the neighborhood radius to best balance the two objectives of discovering real risk clusters and ignoring accidental noise points.

[0034] The calculation process of Euclidean distance, the process of determining the number of clusters using the elbow rule, and the specific process of clustering using the DBSCAN density clustering algorithm are all well-known techniques and will not be elaborated further.

[0035] Furthermore, in this embodiment, based on the welding deformation probability of all mesh nodes in each cluster, the comprehensive deformation risk value of each cluster is determined in the 3D model under each welding fixture layout scheme. Specifically: The average value of the welding deformation probability of all grid nodes in each cluster is used as the comprehensive deformation risk value of each cluster.

[0036] Furthermore, based on the comprehensive deformation risk value and the proportion of all mesh nodes in each cluster among all mesh nodes in the 3D mesh model, the risk level under each welding fixture layout scheme is evaluated, specifically: The risk level of each welding fixture layout scheme is positively correlated with the comprehensive deformation risk value of each cluster and the proportion of all mesh nodes in each cluster in the three-dimensional mesh model.

[0037] It should be understood that a positive correlation means that the dependent variable increases as the independent variable increases, and the dependent variable decreases as the independent variable decreases. The specific relationship can be additive or multiplicative, etc., and is determined by the actual application. This application does not impose any special restrictions.

[0038] Preferably, as one implementation method, in this embodiment, the comprehensive deformation risk value of each cluster is multiplied by the proportion of all grid nodes in the corresponding cluster in all grid nodes in the three-dimensional grid model, and recorded as the product value of each cluster. The sum of the product values ​​of all clusters is used as the risk level under each welding fixture layout scheme.

[0039] Based on the risk level of each welding fixture layout scheme, it can be understood that the risk level is used to characterize the inherent defect risk of each welding fixture layout scheme in suppressing welding deformation. The higher the risk level, the more likely the corresponding layout scheme will cause the final product to have excessive welding deformation, which poses a direct threat to product quality. Among them, the welding deformation probability is the core driving factor of the risk level. If the welding deformation probability is higher, the calculated risk level is higher. This indicates that the corresponding fixture layout scheme cannot effectively constrain the thermal stress caused by the welding process, which indicates that the workpiece will undergo significant geometric deformation at this point, affecting the subsequent assembly and use performance of the fixture, and the corresponding risk level will also increase. Moreover, if the proportion of all grid nodes in the cluster is larger than that of all grid nodes, it acts as a risk amplifier. The larger the proportion, even if the average welding deformation probability of the cluster is not high, it will lead to an increase in the overall risk level due to its wide range of influence, and the corresponding risk level is also larger. Conversely, the lower the probability of welding deformation, the lower the calculated risk level. This indicates that the corresponding fixture layout scheme can effectively match and constrain the thermal stress caused by the welding process, and the geometric deformation of the workpiece in the critical area is well controlled, ensuring the dimensional accuracy of the product and the subsequent assembly performance. The corresponding risk level is also reduced. Furthermore, the smaller the proportion of all grid nodes in the cluster among all grid nodes, the weaker its risk amplification effect. The smaller the proportion, even if there are local high-risk points in the cluster, their contribution to the overall risk level is small due to their limited influence range. This indicates that the problem may only be localized and is easier to solve by fine-tuning the position of individual fixtures. The overall layout scheme still has high reliability, and the corresponding risk level is also low.

[0040] Thus, this embodiment first uses the random forest algorithm to predict the deformation probability of each grid node, then uses DBSCAN clustering to aggregate discrete risk points into spatially meaningful risk clusters, and finally combines the risk intensity and spatial proportion of each cluster to quantify the risk level of each welding fixture layout scheme, thereby transforming the fuzzy deformation problem into a precise and quantifiable evaluation index, providing a scientific basis for fixture layout decisions.

[0041] Step S3: Obtain the 3D ligand model and 3D surface model for each welding fixture layout scheme, and combine the 3D models with a collision detection algorithm to evaluate the collision degree for each welding fixture layout scheme; obtain the path planning data and 3D spatial coordinates of the welding fixture for each welding fixture layout scheme, and use a path planning algorithm to determine the mobility rate for each welding fixture layout scheme; based on the collision degree and the mobility rate, determine the movement characteristic value for each welding fixture layout scheme, and combine the risk degree to determine the welding robustness score for each welding fixture layout scheme.

[0042] Due to the complex spatial coupling relationship of the 3D model of the welding fixture in 3D space, problems such as 3D motion collision and insufficient operating space of the welding fixture are often encountered in the actual welding operation. This makes it impossible for the welding fixture to reach the target position, which in turn makes it impossible for the welding robot to reach it, ultimately resulting in a reduction in welding quality and welding efficiency.

[0043] Therefore, to solve the above problems, a three-dimensional ligand model and a three-dimensional surface model are obtained for each welding fixture layout scheme. Combined with the three-dimensional models, a collision detection algorithm is used to evaluate the collision degree for each welding fixture layout scheme. Path planning data and the three-dimensional spatial coordinates of the welding fixtures are obtained for each welding fixture layout scheme. A path planning algorithm is used to determine the mobility rate for each welding fixture layout scheme. Based on the collision degree and the mobility rate, the movement characteristic value for each welding fixture layout scheme is determined. Combined with the risk degree, a welding robustness score for each welding fixture layout scheme is determined. The specific process is as follows: First, in this embodiment, the three-dimensional surface model and three-dimensional ligand model under each welding fixture layout scheme are exported from CAD software. The three-dimensional surface model is mainly used for efficient collision detection calculation to identify potential spatial interference between the fixture and the tool, and between the fixture and the workpiece. The three-dimensional ligand model includes the complete assembly structure of all fixture components. The process of obtaining the three-dimensional surface model and three-dimensional ligand model under each welding fixture layout scheme using CAD software is a well-known technology and will not be described in detail here.

[0044] Furthermore, the 3D model, 3D ligand model, and 3D surface model under each welding fixture layout scheme are used as inputs to the collision detection algorithm, and the number of collisions of the welding fixture under each welding fixture layout scheme is output. The output number of collisions is recorded as the collision degree under each welding fixture layout scheme.

[0045] It should be noted that there are many commonly used collision detection algorithms. In this embodiment, a bounding box-based collision detection algorithm is used to obtain the number of collisions. That is, the three-dimensional model, three-dimensional ligand model and three-dimensional surface model of the collision degree under each welding fixture layout scheme are used as the input of the bounding box-based collision detection algorithm. The hierarchical bounding box type is set to AABB, the axially aligned bounding box is used, the collision detection accuracy is set to 0.1mm, and the discretization sampling step size is 0.05mm. In practical applications, as other implementation methods, implementers can also use other collision detection algorithms according to specific circumstances. This embodiment does not impose any special restrictions on the selection of collision detection algorithms.

[0046] It should be further explained that in precision mechanical manufacturing and welding fixture design, the positioning tolerances and fit clearances between key components are usually on the order of 0.1mm to 0.5mm. Setting the collision detection accuracy to 0.1mm means that the algorithm can identify interferences that are truly meaningful in engineering. If the accuracy is set too coarsely, many layouts that would lead to assembly difficulties or interference in practice will be incorrectly judged as safe. If the discretization sampling step size is greater than the collision detection accuracy, the collision detection algorithm may detect a safe point at point A and also detect a safe point at point B, but the straight line trajectory between points A and B has actually passed through a thin-walled obstacle, and the collision detection algorithm will completely miss this collision. If the sampling step size is less than or equal to the collision detection accuracy, after the algorithm detects a safe point at point A, the next detection at point A, which is 0.05mm away, will be successfully detected because the distance between point A and the obstacle is less than the collision threshold of 0.1mm.

[0047] The process of counting the number of collisions using a bounding box-based collision detection algorithm is a well-known technique and will not be elaborated further.

[0048] Furthermore, in this embodiment, path planning data for each welding fixture layout scheme is obtained from the process planning module in the CAD software. This includes the starting coordinates of the welding fixture operation path, the three-dimensional coordinates of all welding points, and the position coordinates of the welding fixture in the assembly space for each welding fixture layout scheme. This path planning data defines the spatial path and target position of the welding operation.

[0049] The process of using CAD software to obtain path planning data for each welding fixture layout scheme is a well-known technique and will not be elaborated further.

[0050] Furthermore, this embodiment uses a path planning algorithm to determine the mobility rate of each welding fixture layout scheme based on the path planning data and the three-dimensional spatial coordinates of the welding fixture under each welding fixture layout scheme. Specifically: In this embodiment, the path planning data and the three-dimensional spatial coordinates of the welding fixture under each welding fixture layout scheme are used as inputs to the path planning algorithm, and the optimal path is output. Based on the path planning data for each welding fixture layout scheme, the longest path length is obtained, and the result of dividing the optimal path length by the longest path length is used as the mobility rate for each welding fixture layout scheme.

[0051] The longest path length is calculated based on the coordinates of the starting point, welding point, and target point in the path planning data, ultimately determining the length of the longest route between them.

[0052] It should be noted that there are many commonly used path planning algorithms; this embodiment uses A. The path planning algorithm plans the optimal path, that is, it uses the path planning data and the three-dimensional spatial coordinates of the welding fixtures under each welding fixture layout scheme as A. The input to the path planning algorithm is set with a grid division precision of 0.5 mm and a heuristic function weight coefficient of 1.0. The final output is the optimal path. In practical applications, as another implementation method, the implementer can also use D... This embodiment does not impose any special restrictions on other path planning methods, such as path planning algorithms.

[0053] To clarify, if the mesh precision of path planning is much greater than the collision detection precision, the path planning algorithm will consider two adjacent mesh center points to be safe. However, the straight line segment connecting these two points might pass through a narrow gap of less than 0.5mm but greater than 0.1mm, thus triggering a collision. In this embodiment, the mesh precision of 0.5mm is five times that of the collision detection precision of 0.1mm, providing a reasonable buffer and ensuring that the safety judgment in the path planning stage and the safety judgment in the fine-tuning stage are compatible in scale, thus ensuring the feasibility of the final path. In this embodiment, the heuristic function weight coefficient is set to 1.0 because when the heuristic function weight coefficient is 1.0, it is the standard A... The path planning algorithm can guarantee finding an absolute shortest path from the starting point to the destination. In practical applications, as other implementation methods, implementers can also set their own methods according to specific circumstances. This embodiment does not impose any special restrictions.

[0054] Among them, using A The process of obtaining the optimal path using path planning algorithms is a well-known technique and will not be elaborated further.

[0055] Furthermore, this embodiment determines the movement characteristic value under each welding fixture layout scheme based on the collision degree and the mobility rate, specifically: The movement characteristic value under each welding fixture layout scheme is negatively correlated with the collision degree and mobility rate under each welding fixture layout scheme.

[0056] It should be understood that a negative correlation means that the dependent variable decreases as the independent variable increases, and the dependent variable increases as the independent variable decreases. It can be a subtraction relationship or a division relationship, depending on the actual application.

[0057] Preferably, as one implementation method, in this embodiment, the expression for the movement characteristic value under each welding fixture layout scheme is: In the formula, This represents the movement characteristic value under the i-th welding fixture layout; This represents the collision degree under the i-th welding fixture layout; represents the mobility ratio under the i-th welding fixture layout; exp[ ] represents an exponential function with the natural constant as the base.

[0058] Based on the movement characteristic values ​​under each welding fixture layout scheme, it can be understood that the movement characteristic value is used to characterize the process feasibility level of the welding fixture layout in ensuring unimpeded power and high-efficiency operation. It reflects the quality of the operating space provided by the fixture for welding operations. The higher the movement characteristic value, the freer the fixture moves, the fewer the collisions, and the more guaranteed the efficiency and quality of the welding operation. Among them, the collision degree directly reflects the frequency of interference between the welding fixture and the surrounding environment on the planned path, and has a decisive influence on the movement characteristic value. The greater the collision degree, the more drastically the movement characteristic value drops. This exposes a fatal design flaw in the fixture layout, such as a narrow operating space or no interference, which will directly lead to welding operation interruption, welding fixture damage, or even production stoppage. The corresponding movement characteristic value is actually smaller. At the same time, the mobility rate reflects the economy of the tool path, that is, the detour of the path, and also has a negative impact on the movement characteristic value. The greater the mobility rate, the lower the movement characteristic value. This indicates that the fixture layout forces the tool to take a long detour. Conversely, the smaller the collision degree, the higher the movement characteristic value, indicating that the fixture layout design is reasonable, the spatial interference problem is effectively avoided, and the continuity and safety of welding operations are ensured. The corresponding movement characteristic value is actually larger. At the same time, the smaller the mobility rate, that is, the shorter the path, the higher the movement characteristic value. This indicates that the fixture layout scheme has planned the optimal working path for the tool, reduced ineffective movement, significantly improved production cycle and energy utilization, and achieved high-efficiency production.

[0059] Furthermore, in this embodiment, the movement characteristic values ​​under each welding fixture layout scheme, combined with the aforementioned risk level, are used to determine the welding robustness score under each welding fixture layout scheme. Specifically: The welding robustness score for each welding fixture layout scheme is positively correlated with the movement characteristic value for each welding fixture layout scheme and negatively correlated with the risk level.

[0060] Preferably, the schematic diagram of the welding robustness score extraction process provided in this embodiment is as follows: Figure 2 As shown.

[0061] Preferably, as one implementation, in this embodiment, the sum of the risk level and the preset factor under each welding fixture layout scheme is calculated, and the result of the ratio of the movement characteristic value under each welding fixture layout scheme to the sum is used as the welding robustness score under each welding fixture layout scheme.

[0062] It should be noted that the preset factor is used to prevent the denominator from being 0, and its value is set manually. In this embodiment, the preset factor is 0.01. In actual applications, as other implementation methods, under the premise of ensuring that the denominator is not 0 and does not excessively affect the calculation result, the implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0063] Based on the welding robustness score for each welding fixture layout scheme, it can be understood that the welding robustness score is used to characterize the global optimality of the welding fixture layout scheme after balancing the two core contradictory objectives of suppressing deformation and ensuring accessibility. Among them, the movement eigenvalue reflects the process feasibility of the welding fixture layout scheme and has a direct positive promoting effect on the welding robustness score. The larger the movement eigenvalue, the higher the welding robustness score, indicating that the welding fixture scheme performs well at the operational level and provides reliable physical protection for high-quality welding. Conversely, the smaller the movement eigenvalue, the lower the welding robustness score, indicating that the welding fixture scheme has serious defects at the operational level, such as limited tool movement space, detours, or frequent collisions, resulting in poor process feasibility. Even if deformation can be controlled in theory, it cannot be executed efficiently and smoothly in actual production, thus seriously weakening the overall engineering value of the scheme.

[0064] The risk level reflects the quality and reliability of the welding fixture layout scheme and has a decisive negative constraint on the welding robustness score. The higher the risk level, the lower the welding robustness score, indicating that the welding fixture layout scheme cannot fundamentally guarantee the geometric accuracy of the welded parts and has serious quality risks. Therefore, the corresponding welding robustness score is smaller. Conversely, the lower the risk level, the higher the welding robustness score, indicating that the welding fixture layout scheme can fundamentally and effectively suppress welding deformation, fully guarantee the geometric accuracy and quality reliability of the product, and eliminate the core quality risks. Therefore, the corresponding welding robustness score is also larger.

[0065] Thus far, this embodiment has been completed through A The path planning algorithm plans the optimal path to quantify the mobility rate, and the collision detection algorithm counts the number of collisions. The two are then integrated into a comprehensive movement feature value to characterize the process feasibility. Finally, the movement feature value is combined with the risk level to construct a welding robustness score, thereby realizing a global quantitative evaluation of the welding fixture layout scheme in terms of both quality reliability and process feasibility.

[0066] Step S4: Based on the welding robustness score, obtain a 3D model of the optimal layout of the welding fixture.

[0067] The three-dimensional models under a preset number of welding fixture layout schemes are used as input to the genetic algorithm. The welding robustness score under each welding fixture layout scheme is used as the fitness function value in the genetic algorithm. The number of iterations is set to 100, the crossover probability is set to 0.8, and the mutation probability is set to 0.05. The three-dimensional model under the optimal welding fixture layout is output.

[0068] The preset quantity is set manually. In this embodiment, the preset quantity is 50. In actual application, as other implementation methods, the implementer can also set it according to the specific situation. This embodiment does not impose any special restrictions.

[0069] It should be further explained that in the multi-objective optimization of welding fixture design, the parameter settings of the genetic algorithm are crucial to ensuring that it efficiently finds the global optimum. The number of iterations is set to 100 to ensure that the algorithm has enough generations to converge fully and find a high-quality layout scheme, while controlling the huge computational overhead caused by calling complex simulation and planning algorithms for each evaluation. The crossover probability is set to 0.8 to promote gene combination among different excellent layout schemes and accelerate the evolution of the population towards a better region. The mutation probability is set to 0.05 to prevent the algorithm from getting trapped in local optima too early. These three factors work together to ensure that the final fixture layout scheme achieves the best synergy between welding deformation control and operational accessibility.

[0070] The process of using a genetic algorithm to obtain a 3D model of the optimal layout of the welding fixture is a well-known technique and will not be described in detail here.

[0071] Thus, this embodiment uses the welding robustness score as the fitness function of the genetic algorithm. By iteratively optimizing the fixture position, it ultimately drives the 3D parametric modeling system to automatically generate the optimal fixture layout model with the lowest risk and highest accessibility, realizing an intelligent closed loop from evaluation to design and improving the reliability of welding fixture design.

[0072] It should be noted that the order of the embodiments described above is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some implementations, multitasking and parallel processing are possible or may be advantageous.

[0073] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0074] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them; modifications to the technical solutions described in the foregoing embodiments, or equivalent substitutions of some of the technical features, do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application, and should all be included within the protection scope of this application.

Claims

1. A method for three-dimensional parametric modeling of welding fixtures, characterized in that, The method includes the following steps: Obtain the 3D model of each welding fixture layout scheme and perform mesh processing on it; Analyze the historical welding deformation probability of each grid node to predict the current welding deformation probability of each grid node; cluster all grid nodes, combine the current welding deformation probability of all grid nodes in each cluster, and the proportion of all grid nodes in each cluster in all grid nodes in the 3D mesh model to evaluate the risk level under each welding fixture layout scheme. A three-dimensional ligand model and a three-dimensional surface model are obtained for each welding fixture layout scheme. A collision detection algorithm is then used to evaluate the collision degree for each welding fixture layout scheme, based on the three-dimensional models. Path planning data and the three-dimensional spatial coordinates of the welding fixture are obtained for each welding fixture layout scheme. A path planning algorithm is then used to determine the mobility rate for each welding fixture layout scheme. Based on the collision degree and the mobility rate, the movement characteristic value for each welding fixture layout scheme is determined. Combined with the risk level, a welding robustness score for each welding fixture layout scheme is then determined. Based on the welding robustness score, a 3D model of the optimal layout of the welding fixture is obtained.

2. The method for three-dimensional parametric modeling of welding fixtures as described in claim 1, characterized in that, The method for determining the current welding deformation probability of each grid node is as follows: The random forest algorithm takes all mesh nodes and their historical welding deformation probabilities in the 3D model under each welding fixture layout scheme as input and outputs the current welding deformation probability of each mesh node in the 3D model under each welding fixture layout scheme.

3. The method for three-dimensional parametric modeling of welding fixtures as described in claim 1, characterized in that, The method for determining the distance metric during clustering is as follows: The current 3D coordinates of each grid node are multiplied by a preset first coefficient and the current welding deformation probability is multiplied by a preset second coefficient to form a quadruple. The distance is measured as the Euclidean distance between the grid node quadruples, where the preset second coefficient is greater than the preset first coefficient.

4. The method for three-dimensional parametric modeling of welding fixtures as described in claim 1, characterized in that, The assessment of the risk level under each welding fixture layout scheme includes: In the 3D model of each welding fixture layout scheme, the comprehensive deformation risk value of each cluster is determined based on the welding deformation probability of all grid nodes in each cluster. The risk level of each welding fixture layout scheme is positively correlated with the comprehensive deformation risk value of each cluster and the proportion of all mesh nodes in each cluster in the three-dimensional mesh model.

5. The method for three-dimensional parametric modeling of welding fixtures as described in claim 4, characterized in that, The overall deformation risk value of each cluster is the average of the welding deformation probabilities of all grid nodes in each cluster.

6. The method for three-dimensional parametric modeling of welding fixtures as described in claim 1, characterized in that, The collision degree for each welding fixture layout scheme is the number of collisions of the welding fixture under each welding fixture layout scheme obtained by applying a collision detection algorithm to the three-dimensional model, three-dimensional ligand model and three-dimensional surface model under each welding fixture layout scheme.

7. The method for three-dimensional parametric modeling of welding fixtures as described in claim 1, characterized in that, The method for determining the mobility ratio under each welding fixture layout scheme is as follows: The path planning data and the three-dimensional spatial coordinates of the welding fixture under each welding fixture layout scheme are used as input to the path planning algorithm, and the optimal path is output. Based on the path planning data for each welding fixture layout scheme, the longest path length is obtained, and the result of dividing the optimal path length by the longest path length is used as the mobility rate for each welding fixture layout scheme.

8. The method for three-dimensional parametric modeling of welding fixtures as described in claim 1, characterized in that, The movement characteristic value under each welding fixture layout scheme is negatively correlated with the collision degree and mobility rate under each welding fixture layout scheme.

9. A three-dimensional parametric modeling method for welding fixture design as described in claim 1, characterized in that, The welding robustness score for each welding fixture layout scheme is positively correlated with the movement characteristic value for each welding fixture layout scheme and negatively correlated with the risk level.

10. A three-dimensional parametric modeling method for welding fixture design as described in claim 1, characterized in that, The 3D modeling for obtaining the optimal layout of the welding fixture includes: The three-dimensional models under a preset number of welding fixture layout schemes are used as input to the genetic algorithm. The welding robustness score under each welding fixture layout scheme is used as the fitness function value in the genetic algorithm, and the three-dimensional model under the optimal welding fixture layout is output.

Images