A component topology inheritance and size reconfiguration method for local modification
By determining functional topological anchor points and maintaining boundaries during partial modifications of mechanical products, and reconfiguring dimensional driving relationships, the problems of chain changes in non-target areas and breakage of topological references in the local modification process of existing technologies are solved. This enables flexible dimensional adjustments and stable inheritance of functional structures, improving the efficiency and reliability of component design.
Patent Information
- Application Number
- CN202611363012.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-03
- Publication Date
- 2026-11-17
AI Technical Summary
Existing technologies make it difficult to achieve flexible size adjustments during the partial modification of mechanical products while ensuring that the key functional structures remain unchanged. This leads to unnecessary chain deformations in non-target areas, positioning datum offsets, mating surface failures, or assembly interface misalignments. Furthermore, the lack of explicit protection mechanisms for functional topology relationships results in model reconstruction failures or the need for extensive manual repairs.
By obtaining the parent model of the component to be modified, the feature dependency relationship, topological adjacency relationship and size-driven relationship are analyzed, the functional topological anchor point is determined, and the influence propagation is carried out along the feature dependency relationship and size-driven relationship. The size-driven relationship and size propagation direction are re-determined by using the functional topological anchor point and its boundary truncation propagation path, and local regeneration is performed to keep the topological relationship of the key functional structure unchanged.
It improves the accuracy of controlling the scope of local modifications, maintains the topological inheritance stability of key functional structures, enhances the automatic handling capability of complex dimensional constraint conflicts, takes into account both new design conditions and original design intentions, improves the reuse efficiency of CAD models and the efficiency of local customization, and avoids overall structural damage.
Smart Images

Figure CN123116491A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the field of computer-aided design technology, and in particular to a method for component topology inheritance and size reconfiguration for local modification. Background Technology
[0002] In the serial development and customized production of mechanical products, local modification based on existing mature 3D models is a common design pattern. Designers typically select functionally similar parts from a model library as the parent model and use a computer-aided design system to read its feature tree structure and parameter constraint information. Existing parametric modification techniques mainly rely on preset feature dependencies and dimension-driven chains. When new design conditions such as installation space, interface location, or external boundary are input, the system directly modifies the corresponding dimensional values in the parent model and recalculates the geometric topology according to the original feature generation order and parent-child driving logic, thereby generating a new part model. This process encompasses fully automated processing from sketch updates and feature reconstruction to engineering drawing association, aiming to quickly respond to diverse customer needs.
[0003] However, in practical engineering applications, when faced with complex local modification requirements, existing technologies often struggle to achieve flexible dimensional adjustments while ensuring the critical functional structure remains unchanged. When local design conditions change significantly or multiple constraints conflict, the rigid dimensional-driven logic can easily lead to unnecessary chain deformations in non-target areas, causing positioning datum offsets, mating surface failures, or assembly interface misalignments. Furthermore, due to the lack of explicit protection mechanisms for functional topological relationships, geometric topological splitting or merging caused by dimensional changes often results in the breakage of downstream feature references, leading to model reconstruction failures or requiring extensive manual repairs. This global parameter regeneration method not only destroys proven mature structures but also makes it difficult to automatically reconstruct a reasonable dimensional-driven chain when new and old design conditions conflict, limiting the efficient reuse of existing model resources and the reliability of design iterations. Summary of the Invention
[0004] This application provides a method for component topology inheritance and size reconfiguration for local modification to solve the above problems. The method includes: Obtain the parent model and target design conditions of the component to be modified, and analyze the feature dependencies, topological adjacency relationships and dimension-driven relationships of the parent model; The functional topology anchor points are determined based on the installation interfaces, positioning references, and functional structures in the parent model, and the characteristics of the source of change are determined by comparing the target design conditions with the original design conditions. Starting from the source of change features, the influence propagation proceeds along the feature dependency and size-driven relationships, and terminates when the propagation reaches the preservation boundary corresponding to the functional topology anchor point, in order to determine the local modification influence domain. The dimensional constraint level is determined based on the functional constraint attributes of each dimension within the influence domain of the local modification, and the dimensional constraints corresponding to the target design conditions are added to the original dimensional driving relationship. When a size-driven conflict occurs after the size constraint is added, the size constraint with the higher constraint level is maintained, the original driving relationship of the lower constraint level size that conflicts with it is released, and the driving relationship and size propagation direction of the corresponding size are re-determined to obtain the target size driving relationship. According to the target size driving relationship, the features within the local modification influence domain are solved for size and locally regenerated, while keeping the model structure outside the local modification influence domain and the topological relationship corresponding to the functional topological anchor point unchanged, to obtain the target component model.
[0005] Optionally, the determination of the functional topology anchor point includes: Extract the geometric entities corresponding to the installation interface, positioning reference and functional structure from the parent model, and establish anchor point description data for each geometric entity. The anchor point description data includes at least the geometric entity type, the orientation and positional relationship relative to the reference system, the topological adjacency relationship with adjacent geometric entities and the association relationship with the dimension nodes. Geometric entities with the same functional identifier and whose anchor point description data are related to each other are grouped into a functional topology anchor point, so that the functional topology anchor point is associated with the corresponding functional structure in the parent model through the anchor point description data rather than the geometric entity number.
[0006] Optionally, the preservation boundary corresponding to the functional topology anchor point is determined in the following way: Using the geometric entities contained in the functional topology anchor points as initial entities, extract the geometric entities directly adjacent to the initial entities and the structural features associated with the initial entities through positioning dimensions or mating dimensions; The structural features used to generate the functional topology anchor point, restrict the position of the functional topology anchor point relative to the reference system, or maintain the connection relationship between the functional topology anchor point and adjacent structures are determined as boundary features, and the functional topology anchor point and the boundary features constitute the retention boundary.
[0007] Optionally, the determination of the influence domain of the local modification includes: Using the change source feature as the current affected feature, the next size node is determined along the size-driven relationship based on the change size associated with the current affected feature, and the structural feature corresponding to the next size node is determined as a candidate propagation feature. When the generation parameters of the candidate propagation feature are driven by the previous size node, or when the candidate propagation feature references the geometric entity generated by the previous affected feature, the candidate propagation feature is marked as the affected feature and continues to propagate. When the candidate propagation feature belongs to the functional topology anchor point or the boundary feature, or when further changing the candidate propagation feature will change the directional relationship, positional relationship or associated size corresponding to the functional topology anchor point, the propagation path pointing to the candidate propagation feature is truncated. The affected features and their associated dimensions covered by the untruncation propagation path are defined as the local modification influence domain.
[0008] Optionally, a local size constraint graph is established for the local modification influence domain, wherein the local size constraint graph uses size as size nodes and the driving relationship between one size and another size as directed edges; Each dimension node records at least the current dimension value, allowed variation range, constraint level, and fixed or variable state of the corresponding dimension; each directed edge records at least the driving dimension, the driven dimension, and the geometric constraint relationship between the driving dimension and the driven dimension. The constraint levels, in descending order, include at least the interface locking dimensions directly defined by external assembly conditions, the functional reference dimensions defining positioning or mating relationships, the structural performance dimensions defining structural performance, and the variable dimensions used for local structural adjustments.
[0009] Optionally, the size-driven conflict is determined by the local size constraint diagram, including: After adding the size constraints corresponding to the target design conditions, if it is detected that there are two or more upstream size nodes with definite target values pointing to the same size node and the corresponding solution values are inconsistent, it is determined that there is a multi-source driving conflict. When a directed path is formed after adding the size constraint, starting from any size node and returning to that size node, a closed-loop driving conflict is determined to exist; A path search is performed along the directed edge from the dimension node corresponding to the target design condition. When the search path reaches the dimension node associated with the functional topology anchor point and the target design condition requires the associated dimension to change, an anchor point constraint conflict is determined to exist.
[0010] Optionally, when the size-driven conflict exists, the directed edge causing the size-driven conflict is located, including: The node with the highest constraint level and a defined target value in the conflict path is identified as the baseline driving node. Search upstream dimension nodes in reverse along the direction pointing to the reference driving node until a dimension node in a variable state with a constraint level lower than the reference driving node is found, or a dimension node associated with a functional topology anchor point is found. The directed edges in the reverse search path that are located between the reference driving node and the size node in the variable state, and whose driving direction conflicts with the target size constraint of the reference driving node, are identified as driving edges to be reconstructed.
[0011] Optionally, the re-determination of the driving relationship and dimensional propagation direction for the corresponding dimensions includes: Delete the driving edge to be reconstructed and use the baseline driving node as the reconfiguration starting node; According to the reverse search path, new directed edges opposite to the original driving direction are sequentially established from the reference driving node to the dimension node in the variable state, so that at least one driven dimension in the original dimension chain is converted into a driving dimension and at least one original driving dimension is converted into a driven dimension. When the reconfigured path reaches a dimension node associated with a functional topology anchor point or a dimension node in a fixed state, the creation of the new directed edge is stopped so that the reconfigured dimension propagation path is restricted to the local modification influence domain.
[0012] Optionally, when the dimensional change caused by the target design conditions needs to be absorbed by two or more dimensional nodes in a variable state, the variable dimensional nodes are selected in order of constraint level from low to high. For the currently selected variable-size node, the amount of change that can be absorbed is determined based on the difference between its current size value and the upper or lower limit of the allowable change range, and the portion of the size change that does not exceed the amount of change that can be absorbed is allocated to the current variable-size node. When the size change exceeds the absorbable change of the current variable size node, the current variable size node is adjusted to the boundary value of the corresponding allowable change range, and the remaining size change is further distributed to the variable size node of the next constraint level until the size change is completely absorbed or there is no variable size node that satisfies the allowable change range.
[0013] Optionally, after completing the local regeneration according to the target size-driven relationship, a topology inheritance check is performed on the functional topology anchor points in the regenerated model, including: Candidate geometric entities of the same geometric entity type as each functional topological anchor point in the parent model are extracted from the regenerated model. Based on the orientation and positional relationship of the candidate geometric entities relative to the reference system and the topological adjacency relationship with adjacent geometric entities, they are matched item by item with the anchor point description data of the corresponding functional topological anchor points. Candidate geometric entities that simultaneously satisfy geometric entity type, baseline relationship, and topological adjacency relationship are determined as the inheriting entities of the corresponding functional topological anchor points; When any functional topology anchor does not have a corresponding inheriting entity, the size propagation path connected to that functional topology anchor is determined as a failure path, and the size-driven relationship reconfiguration step is returned to reallocate the size change amount of the variable size nodes in the failure path.
[0014] Through the above technical solution, this application has the following beneficial effects: This application improves the accuracy of controlling the scope of local modifications. Starting with the characteristics of the source of change, it combines feature dependencies and size-driven relationships to propagate the impact, and utilizes functional topology anchors and their holding boundaries to truncate the propagation path. This restricts the modification operation to the local range required to meet the target design conditions, reducing unnecessary changes in non-target areas caused by parameter linkages.
[0015] Improve the stability of topological inheritance of key functional structures. By setting installation interfaces, positioning references, and key functional structures as functional topological anchor points, and using descriptive data such as geometric entity types, reference relationships, and topological adjacency relationships to characterize the anchor points, rather than simply relying on the geometric entity numbers within the CAD system, the risk of reference breakage caused by surface splitting, merging, or renumbering during local modifications can be reduced, which is conducive to maintaining the stability of key installation, positioning, and mating relationships.
[0016] This application improves the automatic handling capability of complex dimensional constraint conflicts. It sets corresponding constraint levels based on the functional attributes of different dimensions, and when new design conditions conflict with the original dimensional driving relationships, it prioritizes maintaining higher-level constraints while simultaneously releasing or reconstructing the driving relationships of lower-level dimensions. This allows the original fixed dimensional driving chain to be adjusted according to new design requirements, reducing dimensional solution failures caused by conflicts between multi-source drives, closed-loop drives, and anchor point constraints.
[0017] It balances new design conditions with original design intent. By redefining the size-driven relationships and size propagation directions, and only solving for and regenerating the dimensions of features within the influence domain of local modifications, it is possible to meet new installation space, interface location, or size requirements while keeping the topological relationships of the external structure and functional topological anchor points essentially unchanged. This avoids overall structural damage to mature component models due to local size adjustments.
[0018] This improves the engineering rationality and feasibility of the dimensional reconfiguration results. For dimensions that need adjustment, this application further combines the allowable range of dimensional changes, constraint levels, and dimensional deviation costs for local solutions. This allows dimensional changes to be absorbed primarily by dimensions with lower constraints and adjustment margins. While ensuring key dimensional conditions, this minimizes the dimensional deviation of the target model relative to the parent model, which is beneficial for obtaining reconfiguration results that are more consistent with the original design basis.
[0019] This application improves the reuse efficiency and local customization efficiency of existing CAD models. Overall, it addresses issues such as chain changes in non-target regions, rigidity of dimension drives, and broken topological references that easily occur during existing parametric modification processes through a collaborative processing approach involving functional topology anchor point protection, local influence domain determination, hierarchical dimensional constraints, reconfiguration of driving relationships, and local regeneration. This enables mature master models to adapt to new design conditions while maintaining their core functional structure, improving the efficiency and reliability of serialized design and customized modification of components. Attached Figure Description
[0020] To more clearly illustrate the technical solutions 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0021] Figure 1 This is a flowchart illustrating a component topology inheritance and size reconfiguration method for local modification, provided as an embodiment of this application.
[0022] Figure 2 This is a schematic diagram of the functional topology anchor point, the maintaining boundary, and the influence domain of local modification provided in an embodiment of this application.
[0023] Figure 3 This is a partial dimensional constraint diagram and a schematic diagram of dimensional driving conflicts provided for an embodiment of this application.
[0024] Figure 4 This is a schematic diagram of size-driven relationship reconfiguration and size change allocation provided in an embodiment of this application.
[0025] Figure 5 This is a schematic diagram of local regeneration and functional topology anchor point inheritance verification provided in an embodiment of this application. Detailed Implementation
[0026] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without creative effort are within the scope of protection of this application.
[0027] Furthermore, the term "and / or" in this article is merely a description of the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A existing alone, A and B existing simultaneously, or B existing alone. Additionally, the character " / " in this article, unless otherwise specified, generally indicates that the preceding and following related objects have an "or" relationship.
[0028] The embodiments of this application will now be described in further detail with reference to the accompanying drawings.
[0029] Example 1 like Figure 1 The diagram shown is a flowchart of a component topology inheritance and size reconfiguration method for local modification provided in an embodiment of this application. The method includes: Step S101: Obtain the parent model and target design conditions of the part to be modified, and analyze the feature dependency relationship, topological adjacency relationship and dimension-driven relationship of the parent model.
[0030] This step aims to construct a basic data model and associated mappings to support local modifications. The parent model is a 3D CAD model selected from an existing component model library that belongs to the same functional category or has the same main interface structure as the target component. When selecting the parent model, priority is given to ensuring consistency in basic functions, main installation interface types, and positioning reference systems. Target design conditions may refer to new installation space constraints, interface position adjustment parameters, or functional dimension change requirements proposed by the customer or downstream engineering, such as new mounting hole center distances, shortened outer contour lengths, or adjusted mating surface positions. The parsing process is completed by reading the feature tree, geometric topology data, and dimensional constraint data within the parent model. Among them, feature dependencies describe the parent-child and reference relationships of each feature in the model, such as sketches depending on datum planes and bosses depending on sketches; topological adjacency relationships record the contact and connection of geometric entities (faces, edges, points) in space, such as face-to-face contact and edge-to-edge overlap, and this relationship is independent of the construction order of the feature tree; dimension-driven relationships define the logical associations between dimensional parameters, that is, the rules that drive the linked updates of other related dimensions when one dimension changes. By analyzing the three relationships mentioned above, a complete association graph covering geometric structure and parameter constraints is constructed, transforming the parent model from a simple geometric object into a structured data object that can be logically analyzed.
[0031] Step S102: Determine the functional topology anchor point based on the installation interface, positioning datum and functional structure in the parent model, and determine the source characteristics of change by comparing the target design conditions with the original design conditions.
[0032] This step is used to define the core functions that must remain unchanged during the modification process and the starting point for triggering the modification. Functional topology anchors refer to key structures that must maintain their functional meaning or relative topological relationship during this local modification process, embodying the core assembly and design intent of the components. Functional topology anchors typically include three categories: the first category is assembly positioning anchors, such as main mounting reference surfaces, positioning holes, positioning pin holes, positioning grooves, and central axes; the second category is functional interface anchors, such as threaded holes, flange connection surfaces, pipe interfaces, electrical connection holes, and sealing contact surfaces that connect with other parts; the third category is critical functional anchors, such as bearing mounting surfaces, guide surfaces, kinematic mating surfaces, sealing grooves, and critical load-bearing support surfaces. To prevent reference failure due to changes in Face ID or Edge ID within the CAD system after model changes, functional topology anchors are defined by descriptive data including geometric type, positional relationship relative to the reference system, topological adjacency relationship with adjacent geometric entities, and the functional label they bear, rather than relying on a single geometric number. Variation source features can refer to dimensions or features directly controlled by the new design conditions, which are determined by comparing the target design conditions with the original design conditions of the parent model. The comparison process forms a set of differences, each of which includes the original conditions, the new conditions, the corresponding design parameters, and related features. For example, when the original mounting hole center distance is 120mm and the new design requirement is 135mm, the dimensions and hole features directly related to that hole center distance are identified as variation source features.
[0033] Step S103: Starting from the source feature of the change, propagate the influence along the feature dependency relationship and the size-driven relationship, and terminate the propagation when it reaches the preservation boundary corresponding to the functional topology anchor point, so as to determine the local modification influence domain.
[0034] This step strictly limits the scope of modification to the minimum area necessary to meet the new design conditions, avoiding unnecessary damage to mature functional structures. Influence propagation refers to the process of spreading analysis from the source feature node along both the feature-dependent direction and the size-driven direction to surrounding related nodes. During propagation, for each node to be propagated, the system determines in real time whether it meets the propagation termination conditions. The maintenance boundary is a virtual barrier composed of functional topology anchors and their associated protective structures; propagation stops when the path reaches this boundary. Specific propagation termination conditions include: the current node is an immutable functional topology anchor; the current node is a protective structure directly associated with the functional topology anchor, such as a reference feature used to generate the anchor or a positioning dimension that maintains the anchor's position; the current dimension is a locked functional dimension; or changes to the current node will not affect the satisfaction of the new design conditions. Finally, all affected features and their associated dimensions covered by all uninterrupted propagation paths are determined as the local modification influence domain. This influence domain constitutes a local subgraph of the original model. Subsequent size reconfiguration and feature regeneration are performed only on this subgraph, while structures outside this subgraph retain the original geometric and topological relationships of the parent model, thereby achieving precise control with minimal modifications.
[0035] Step S104: Determine the dimensional constraint level based on the functional constraint attributes of each dimension within the local modification influence domain, and add the dimensional constraints corresponding to the target design conditions to the original dimensional driving relationship.
[0036] This step aims to establish priority standards for dimensional conflict arbitration and inject new design requirements into the existing parameter system. Dimensional constraint levels are based on the importance of dimensions within the influence domain of local modifications according to the design function they perform. Following a descending order of constraint strength, dimensional constraint levels include at least: Level 1 constraints, which are interface locking dimensions directly defined by external assembly conditions, such as mounting hole distances, flange diameters, interface center positions, and external connection heights. These dimensions are generally prohibited from being altered through internal dimensional adjustments. Level 2 constraints, which are functional reference dimensions defining positioning or mating relationships, such as the distance from positioning holes to reference surfaces, coaxial positions of bearing holes, sealing surface positions, and guide surface distances. Level 3 constraints, which are structural performance dimensions defining structural performance, such as wall thickness, stiffener thickness, minimum fillet thickness, and boss thickness to ensure part strength or stiffness. Level 4 constraints, which are variable dimensions used for local structural adjustments, such as local step lengths, auxiliary chamfers, non-functional fillets, and local contour distances. Adding the dimensional constraints corresponding to the target design conditions to the original dimension-driven relationship can mean converting the specific values in the new design conditions into target constraint values and writing them into the corresponding change source dimension nodes, making them new constraints in the local dimension-driven diagram, thereby triggering the re-solution of the original dimension chain.
[0037] Step S105: When a size-driven conflict occurs after adding a size constraint, maintain the size constraint with the higher constraint level, release the original driving relationship of the lower constraint level size that conflicts with it, and redetermine the driving relationship and size propagation direction of the corresponding size to obtain the target size driving relationship.
[0038] This step resolves parameter inconsistencies caused by differences between old and new design conditions by dynamically reconstructing the dimensional logic. Dimension-driven conflicts refer to situations where, after the addition of new constraints, the local dimensional drive diagram cannot be directly satisfied through numerical updates. These mainly include multiple drive conflicts where the same dimension is controlled by two different upstream dimensions, closed-loop drive conflicts where multiple dimensions form a closed drive loop, and lock dimension change conflicts where changes in lower-level dimensions require changes in higher-level locked dimensions. When such conflicts are detected, the system handles them according to the principle of maintaining high priority and yielding to low priority. Specific operations include: if the constraint level of the original driving dimension is lower than the newly added design constraint, the driving attribute of the original dimension is released and it is converted into a driven dimension; subsequently, the driving relationship is re-established according to the new constraint level, for example, reversing the driving direction from determining the mounting hole position based on the outer contour to determining the adjustable outer contour dimension from the mounting hole position; if a dimensional chain crosses a functional topology anchor point, the original dimensional chain is split into an affected first local dimensional chain and an unchanged second local dimensional chain, using the dimensional node corresponding to the functional topology anchor point as the dividing point. Through this series of reconfiguration operations, a conflict-free target dimensional drive relationship suitable for the target component is finally formed.
[0039] Step S106: Solve for the dimensions of the features within the local modification influence domain and regenerate them locally according to the target size driving relationship, while keeping the topological relationship corresponding to the model structure and functional topological anchor points outside the local modification influence domain unchanged, to obtain the target component model.
[0040] This step is the execution stage for finally generating a solid model that meets the new design conditions and retains the original functional intent. Dimensioning based on the target size-driven relationship can refer to using a reconfigured local size-driven diagram to determine the specific geometric parameter values of each feature within the local modification influence domain through analytical equations or iterative calculations. Local regeneration can refer to performing update operations only on features within the local modification influence domain in the CAD system. During regeneration, the system retains all model structures outside the modification influence domain, using the topological relationship between the boundary of the modification influence domain and the functional topological anchor points as fixed boundary conditions. Within these fixed boundaries, the system recalculates and generates the geometry of the local structure. For example, if a mounting boss needs to be lengthened, but the mounting hole on the boss is a functional topological anchor point, then during regeneration, the axis of the mounting hole and its relationship relative to the datum plane remain unchanged; only the outer contour of the boss or the edge fillets are allowed to be recalculated. This approach ensures that the target component model strictly inherits the key functional topological relationships in the parent model while meeting the new target design conditions, avoiding overall model failure or assembly interface misalignment due to local modifications.
[0041] To ensure the above dimensional solution process has executable numerical boundaries, the dimensions involved in the adjustment within the local modification influence domain can be grouped into a dimension vector x. Under the premise of satisfying the target dimension driving relationship, target design conditions, and functional topology anchor point fixing conditions, the solution objective is to minimize the weighted deviation of the dimensions relative to the original dimensions of the parent model. The corresponding local dimension constraint solution can be expressed as: ; In the formula, The dimension vectors involved in the dimension solution within the influence domain of the local modification; The target size vector obtained by solving; The set of size nodes that are allowed to participate in the adjustment; For the first One dimension to be determined. This is the original dimension value of the size in the parent model; and The first The lower and upper limits of the permissible variation range for each size; The dimension deviation cost coefficient can be set according to the existing dimension constraint level, so that dimensions with higher constraint levels have greater deviation costs. To complete the target set of directed edges after the reconfiguration of the driving relationship; For directed edges The constraint equations corresponding to the recorded geometric constraint relationships; This refers to the set of nodes whose dimensions are directly given by the target design conditions. For the corresponding target size value; This is a set of fixed-size nodes associated with functional topology anchor points. The dimensions associated with functional topology anchor points are directly maintained using equality constraints, without approximating them by increasing weights. For linear geometric constraints, they can be solved directly using a system of linear equations; for cases with nonlinear geometric constraints, the CAD parameter solver can be invoked for iteration, ending the iteration when the residuals of each constraint equation are no greater than the solution tolerance set by the CAD system.
[0042] The aforementioned technical process locks in key design intent through functional topology anchors, defines the minimum modification range using influence propagation, and dynamically reconstructs the driving logic based on dimensional constraint levels. This collaboratively solves the unsolvable problems of chain changes in non-target areas and dimensional conflicts encountered in the local modification of mechanical parts. Functional topology anchors, as the core of semantic anchoring, effectively block the ineffective propagation of dimensional changes, ensuring the stability of the installation interface and positioning reference. The reconfiguration of dimensional driving relationships breaks the rigid constraints of the original parameter chain, enabling the model to flexibly adapt to changes in external design conditions. Thus, while maintaining the core functionality, it achieves efficient and precise customization of parts.
[0043] Example 2 In one optional implementation, a process for determining functional topology anchor points is provided for embodiments of this application, including: Step S21: Extract the geometric entities corresponding to the installation interface, positioning datum, and functional structure from the parent model. Establish anchor point description data for each geometric entity. The anchor point description data shall include at least the geometric entity type, the orientation and positional relationship relative to the datum system, the topological adjacency relationship with adjacent geometric entities, and the association relationship with the dimension nodes.
[0044] By constructing high-dimensional topological description information to replace traditional single entity numbers, the problem of reference invalidation caused by surface splitting, merging, or reconstruction during model modification can be solved. First, the feature tree and geometric topology of the parent model are traversed to identify geometric entities that conform to the definitions of installation interfaces, positioning datums, and functional structures. Geometric entities include, but are not limited to, datum surfaces, datum axes, cylindrical surfaces, planes, hole features, slot features, or mating surfaces, which constitute the physical basis for component assembly and functional implementation. For each extracted geometric entity, corresponding anchor point description data is generated. This anchor point description data, as a structured fingerprint that uniquely represents the physical form and logical position of the entity in model space, contains information in the following four dimensions: First, the geometric entity type defines the physical morphology classification of the entity, such as determining whether the entity is a plane, cylinder, sphere, or complex composite surface. This classification ensures the consistency of functional semantics during model variation. Second, the orientation and positional relationship relative to the datum system are used to describe the spatial pose of the entity in the global coordinate system or local assembly coordinate system. By calculating parameters such as the angle between the entity's normal vector and the datum axis, and the distance from key points to the datum plane, an absolute positional description independent of the model's historical construction order is constructed. Third, the topological adjacency relationship with adjacent geometric entities records the first-level connection environment around the entity. For example, whether a cylindrical hole entrance surface is adjacent to the top surface of a boss or a bottom plate plane, a snapshot of the local topological environment is formed by recording the types of adjacent entities and the geometric properties of the connection boundaries. Fourth, the association relationship with dimension nodes binds the geometric entity to design parameters, recording the dimensional constraints that directly control the shape or position of the entity, thereby enabling rapid tracing of affected geometric objects during dimension-driven relationship reconfiguration. The above four data items together constitute the anchor point description data, which allows the system to match the same functional structure through this data, even if the Face ID or Edge ID inside the CAD system changes, as long as the physical existence of the geometric entity and its relationship with the surrounding environment do not change fundamentally.
[0045] Step S22: Combine geometric entities with the same functional identifier and whose anchor point description data are related into a functional topology anchor point, so that the functional topology anchor point is associated with the corresponding functional structure in the parent model through the anchor point description data rather than the geometric entity number.
[0046] By aggregating discrete geometric entities into functional units with clear engineering semantics, an abstract mapping from geometric data to functional objects is achieved. The anchor point description data generated in the preceding steps is not only used for the identification of individual entities but also serves as the basis for cluster analysis. In the specific execution process, key structures in the parent model are first assigned functional identifiers. These identifiers can be, for example, primary positioning datums, flange connection interfaces, sealing mating surfaces, or bearing mounting holes, reflecting the specific engineering roles of components in the assembly. Subsequently, a set of geometric entities with the same functional identifiers is selected, and the inherent logical relationships between the anchor point description data of each geometric entity within this set are further analyzed.
[0047] When the anchor point description data of multiple geometric entities are detected to exhibit continuous, compliant, or inclusive relationships in space, these geometric entities are logically combined into a whole, namely a functional topology anchor point. For example, the axial direction of a cylindrical entity and the normal vector of a planar entity are perpendicular, and their positional distance meets a preset compliance tolerance; or, in terms of topological adjacency, the boundary of entity A is part of entity B. This combination method breaks the rigid constraints of the parent-child relationship based on feature trees in traditional CAD models, allowing the aggregation of physically cooperating geometric elements across different feature levels. The final generated functional topology anchor point stores the above-mentioned aggregated anchor point description data set, rather than a single internal system number. This means that no matter how subsequent local modification operations change the construction order of the feature tree or cause the geometric entities to be renumbered, as long as the physical topology of the functional area remains stable, the functional topology anchor point can be accurately retrieved and reused based on the anchor point description data, thus providing a stable data foundation for locking and inheriting key functional areas during local modification.
[0048] Through the above steps, an anchor point definition mechanism based on topological semantics rather than temporary IDs is constructed, which ensures that when the model undergoes local changes, the key installation interfaces and positioning benchmarks can be stably identified and maintained, providing a reliable reference system for the boundary delineation of the subsequent size influence domain.
[0049] Example 3 One possible implementation is, such as Figure 2 As shown, this application provides a method for determining the preservation boundary corresponding to a functional topology anchor point. This method is also used to specifically define the scope that needs to be frozen and protected during local modifications, ensuring that the modification operation does not damage the stability of the core functional structure.
[0050] Step S301: Using the geometric entities contained in the functional topology anchor point as the initial entity, extract the geometric entities directly adjacent to the initial entity and the structural features associated with the initial entity through positioning dimensions or mating dimensions.
[0051] Functional topology anchors are core structural units that must maintain functional integrity during the modification process. Their corresponding preservation boundaries are not simple geometric buffers, but rather functional protective circles formed by adjacent structures that directly support the existence and positioning of the anchors. When constructing this preservation boundary, the geometric entities of the functional topology anchors themselves are first locked as the initial entities for the initial search. These initial entities are usually represented by specific geometric shapes such as cylindrical surfaces, planes, and axes.
[0052] Based on the defined initial entity, two aspects of association extraction are performed. First, based on the connectivity of the geometric topology, directly adjacent geometric entities that share a boundary or have direct physical contact with the initial entity in three-dimensional space are retrieved, such as the top surface of a boss connected to the inlet face of a cylindrical hole or a side edge sharing an edge with a reference surface. Second, based on the dimensional constraint network of the parametric model, structural features that establish logical associations with the initial entity through positioning dimensions or mating dimensions are found. Positioning dimensions define the position of the entity relative to the coordinate system or reference, while mating dimensions define the relative clearance or interference between two assembled entities.
[0053] This dual search, combining geometric adjacency and dimensional association, comprehensively covers surrounding structures that physically or logically directly support functional topological anchors. For example, when the initial entity is a cylindrical surface with a positioning hole, the extracted content includes not only the hole wall features directly surrounding the cylindrical surface, but also the shaft-like feature contours associated with the hole through mating dimensions, and the base plate edge features whose center distance of the hole is determined by the positioning dimensions.
[0054] Step S302: The structural features used to generate functional topology anchors, restrict the position of functional topology anchors relative to the reference system, or maintain the connection relationship between functional topology anchors and adjacent structures are determined as boundary features, and the functional topology anchors and boundary features constitute the maintenance boundary.
[0055] After extracting the surrounding potentially relevant features, the supporting functions of the functional topology anchors are strictly screened based on each structural feature. The core criterion for screening is whether the feature undertakes any one of the three key functions: generating anchors, restricting anchor positions, or maintaining anchor connections.
[0056] Features used to generate functional topology anchors can refer to parent features that form the basis of the anchor's geometry, such as the extruded sketch profile for generating a cylindrical hole feature or the placement surface for placing a hole feature. These features directly determine the existence of the anchor. Features that restrict the position of a functional topology anchor relative to the datum system can refer to features that lock the anchor in a specific location in the design space through dimensional constraints, such as datum plane features or reference plane features associated with the positioning dimension from the master datum plane to the anchor's central axis. These features ensure the accuracy of the anchor's positioning. Features that maintain the connection between the functional topology anchor and adjacent structures can refer to features that ensure the normal assembly and force transmission of the anchor with other components or the main structure, such as the flange side or welded contact surface that mates with the anchor. These features ensure the reliability of the connection.
[0057] After identifying features that meet any of the above conditions as boundary features, these boundary features, together with the core functional topological anchor points, form a closed or semi-closed set. This set constitutes the preservation boundary ultimately used to terminate the propagation of the modification's impact. This boundary construction method based on functional semantics ensures that during local modification, only structures located outside this preservation boundary are allowed to be modified or reconstructed, while the anchor points and their key supporting structures within the boundary are strictly locked. This allows for flexible local design while maximizing the inheritance and protection of the original model's core design intent.
[0058] Among the aforementioned technical features, defining structural and boundary features is a prerequisite for ensuring that modifications only affect non-critical areas. Structural features, as constituent units of the parent model, are endowed with specific protective value through their topological or dimensional relationships with functional topological anchors. Boundary features, further refined from structural features, specifically refer to the subset that plays a decisive supporting role in the function of the anchors. Their synergistic effect ensures that the boundary is not merely a geometric range, but a dynamic barrier with a clearly defined functional protection logic. When the impact of modifications propagates to this boundary, the propagation path is effectively blocked because the boundary features are protected, thus avoiding assembly failure or functional loss due to erroneous modifications to critical structures. This mechanism provides a safe and stable implementation environment for subsequent local reconfiguration of dimension-driven relationships.
[0059] Example 4 like Figure 2 As shown, in one optional implementation, a process for determining the influence domain of a local modification is provided for embodiments of this application. This method further includes: Step S401: Using the change source feature as the current affected feature, determine the next size node along the size-driven relationship based on the change size associated with the current affected feature, and determine the structural feature corresponding to the next size node as the candidate propagation feature.
[0060] The source of change feature can be a feature that, after comparing the target design conditions with the original design conditions, is directly determined to require modification, such as a boss feature that must be displaced due to adjustment of mounting hole positions. Starting from this source of change feature, the dimensional parameters directly associated with it are retrieved as the changed dimensions. Using the dimension-driven graph resolved from the parent model, the downstream dimensional nodes controlled by this changed dimension are traced. The structural feature to which each identified next dimensional node is attached or described is marked as a candidate propagation feature. This process transforms abstract design changes into concrete geometric feature traversal actions, giving the propagation path a clear topological and parametric basis. For example, if the source of change feature is the outer contour of the bracket, and its length dimension changes, driving the position dimension of the reinforcing rib, then the reinforcing rib feature is determined as the next-level candidate propagation feature.
[0061] Step S402: When the generation parameters of a candidate propagation feature are driven by the previous size node, or when the candidate propagation feature references the geometric entity generated by the previous affected feature, mark the candidate propagation feature as the affected feature and continue propagating.
[0062] For initially identified candidate propagation features, a dual-condition determination is used to determine whether they are truly affected by dimensional fluctuations, thus avoiding the erroneous inclusion of geometrically adjacent but parameter-independent features in the influence domain. On one hand, it checks whether the generation parameters of the candidate propagation feature (such as extrusion depth, rotation angle, sketch positioning dimensions, etc.) are directly driven by the values of the previous dimension node; on the other hand, it checks whether the candidate propagation feature references the geometric entities generated by the previously affected feature during the modeling process (such as referencing the sketch outline or face of the previous feature as a positioning reference). If either of these conditions is met, it indicates that the shape or position of the candidate propagation feature will change in tandem with the changes in the preceding steps, and therefore it is marked as an affected feature. Subsequently, using this affected feature as the new propagation starting point, the search continues downstream along the dimension-driven relationship until all associated paths are traversed or truncated. This determination logic based on parameter dependence and entity references, compared to simple adjacency search, can more accurately reflect the reconstruction order and constraint transmission relationships within the CAD model.
[0063] Step S403: When the candidate propagation feature belongs to the functional topology anchor point or boundary feature, or when further changing the candidate propagation feature will change the directional relationship, positional relationship or associated size of the functional topology anchor point, the propagation path pointing to the candidate propagation feature is truncated.
[0064] Functional topology anchors and boundary features represent design intentions that must remain unchanged during modification, such as mounting reference surfaces and positioning holes. During propagation, the attributes of candidate propagation features and their relationship to functional topology anchors are monitored in real time. If a candidate propagation feature itself belongs to the set of functional topology anchors or the set of boundary features it constitutes, propagation in that direction is immediately stopped, and the feature's state is locked. Furthermore, even if a candidate propagation feature does not belong to an anchor point itself, a predictive analysis is performed: if further changes to the candidate propagation feature would cause deviations in the directional relationships (e.g., axis perpendicularity), positional relationships (e.g., hole distance reference), or associated dimensions (e.g., mating surface distance) of the functional topology anchors, then this feature is considered a support feature within the protected area, and the propagation path is also cut off. This dual-cutoff mechanism ensures that the impact of local modifications is strictly limited to non-core functional areas, thereby maintaining the stability of product assembly and function.
[0065] To determine whether further changes to the candidate propagation features will affect the prediction of the functional topology anchor point, a size trial calculation can be performed on the candidate propagation features without writing back to the parent model. This yields the predicted direction vector, predicted position vector, and predicted associated size of the functional topology anchor point a, and the direction deviation, position deviation, and associated size deviation are calculated. The following normalized deviation criterion can be used: ; In the formula, and These represent the functional topology anchor points in the parent model after trial calculation. The normalized unit direction vector; and These represent the position vectors of the anchor point relative to the reference system in the parent model after the trial calculation; and These represent the first and second anchor points associated with the parent model after trial calculation. One size; , and These are, respectively, directional deviation, positional deviation, and maximum associated dimensional deviation; , and These are the corresponding allowable directional deviation, allowable positional deviation, and allowable dimensional deviation; This is the normalized anchor point deviation index. When... When the value is greater than 1, it indicates that at least one anchor point preservation condition will exceed the allowed range, and the propagation path pointing to the current candidate propagation feature will be truncated; when If the value is ≤1, the impact propagation along that path is allowed to continue. , and The tolerance for the anchor point can be determined based on the assembly angle tolerance, positioning / position tolerance, related dimension design tolerance, and CAD solution tolerance, with the most stringent constraint being preferred. For anchor points with the same functional significance in both directions of the axis, the absolute value of the dot product of the direction vectors is taken when calculating the direction deviation to obtain the equivalent minimum included angle.
[0066] Step S404: The affected features and their associated dimensions covered by the untruncation propagation path are determined as the local modification influence domain.
[0067] After the propagation and filtering steps described above, a directed graph is formed in the system containing all feature nodes on paths originating from the source features and not blocked by functional topological anchors or boundary features. All affected features covered by these uninterrupted propagation paths are collected, and all dimensional parameters associated with these features are taken into consideration, collectively defining the local modification influence domain. This influence domain not only lists the geometric features that need to be regenerated but also implicitly contains the local dimensional chains required to support the regeneration of these features. Since branches protected by functional topological anchors have been excluded during propagation, the determined local modification influence domain is the smallest subset of features that satisfies the target design conditions. Subsequent dimensional-driven relationship reconfiguration and model regeneration operations are only performed within this domain, while the model structure outside the influence domain remains unchanged, thus achieving efficient and safe local modification.
[0068] Example 5 In yet another alternative embodiment, such as Figure 3 As shown, the method further includes: establishing a local dimension constraint diagram for the local modification influence domain. The local dimension constraint diagram uses dimensions as dimension nodes and the driving relationship between one dimension and another dimension as directed edges. Each dimension node records at least the current dimension value, allowable variation range, constraint level, and fixed or variable state of the corresponding dimension. Each directed edge records at least the driving dimension, the driven dimension, and the geometric constraint relationship between the driving dimension and the driven dimension. The constraint levels, in descending order, include at least the interface locking dimensions directly defined by external assembly conditions, the functional reference dimensions that define positioning or mating relationships, the structural performance dimensions that define structural performance, and the variable dimensions used for local structural adjustments.
[0069] A local size constraint graph is established for the local modification influence domain. This local size constraint graph is the core data structure used to express and manage the logical relationships of dimensions within the modification region; its essence is a directed graph model. This graph is not a simple copy of all size relationships in the parent model, but only covers the set of dimensions that interact within the local modification influence domain defined in the above embodiments. By abstracting dimensions as nodes in the graph and the driving and dependency relationships between dimensions as directed edges connecting nodes, the discrete size constraints originally implicit in feature trees or parametric equations can be transformed into a structured network that can be traversed, analyzed, and computed by computer algorithms. This graph-structured representation provides a clear mathematical foundation for subsequent size conflict detection, driving chain tracing, and constraint solving processes.
[0070] Local dimensional constraint diagrams use dimensions as dimension nodes and directed edges representing the driving relationship between one dimension and another. In the graph structure, dimension nodes represent specific design parameters within the influence domain of a local modification, such as the length of a boss, the center distance of a hole, or the wall thickness. Directed edges characterize the causal dependency flow between dimension nodes; that is, when the driving dimension, as the starting point, changes, the driven dimension, as the ending point, needs to be adaptively adjusted accordingly. The directionality of the directed edges defines the parameter transmission path in the design intent. For example, if the total length dimension node points to the hole location dimension node via a directed edge, it indicates that the hole location is determined by the total length. This explicit directional record provides direct data for identifying dimension driving conflicts (such as multiple driving forces or closed-loop dependencies) in subsequent steps.
[0071] Each dimension node records at least the current dimension value, allowable variation range, constraint level, and fixed or variable state for the corresponding dimension. The attribute information stored in the dimension node constitutes a multi-dimensional vector for conflict resolution and reconfiguration decisions. The current dimension value records the initial value of the dimension in the parent model; the allowable variation range defines the upper and lower limits of the dimension's possible value while satisfying engineering rationality (such as avoiding interference and ensuring minimum wall thickness); the constraint level indicates the importance and immutability of the dimension in the design; and the fixed or variable state indicates whether the dimension is allowed to be modified in the current modification task. Through the quantified recording of these attributes, the dimension node is no longer a simple numerical variable, but an intelligent entity carrying rich engineering semantics, enabling the algorithm to automatically determine which dimensions should remain unchanged and which dimensions have room for adjustment when a conflict occurs.
[0072] Each directed edge must record at least the driving dimension, the driven dimension, and the corresponding geometric constraints between them. Besides the directionality of the connecting nodes, the directed edge must also explicitly record the specific geometric relationship between the driving and driven dimensions. Geometric constraints describe the interaction mechanism between the two dimensions in geometric space, such as parallel distance, coaxial positioning, angular constraints, or coincidence. Recording specific geometric constraints is a necessary condition to ensure that the dimensional solution is physically achievable. For example, when the driving dimension represents the position of a plane and the driven dimension represents the center distance of a hole, the geometric constraint between them can be recorded as the hole center being located in the normal direction of the plane at a specific distance. This detailed recording of geometric relationships allows for accurate assessment of the actual impact of dimensional changes on the model's geometry when reconfiguring the driving relationships.
[0073] The constraint levels, in descending order, include at least interface locking dimensions directly defined by external assembly conditions, functional reference dimensions defining positioning or mating relationships, structural performance dimensions defining structural performance, and variable dimensions used for local structural adjustments. This constraint level system serves as the decision-making benchmark for dimensional conflict arbitration and driving relationship reconfiguration in this application's embodiments. Its classification is based on the functional weight of dimensions in engineering applications, rather than geometric attributes. Interface locking dimensions are at the highest level; these dimensions are directly determined by the interfaces of external connectors or customer equipment and are generally prohibited from being changed. Functional reference dimensions are next, involving core functional attributes such as component positioning references and mating accuracy. Structural performance dimensions are next, primarily focusing on physical performance requirements such as strength and stiffness. Variable dimensions are at the lowest level, typically referring to local structural dimensions that only serve a transitional or decorative function, possessing the greatest degree of adjustment freedom. This hierarchical constraint level definition provides a clear execution standard for the strategy of maintaining high-level constraints and adjusting low-level constraints in subsequent steps, ensuring that the modification results prioritize meeting key functional requirements and interface compatibility.
[0074] Example 6 In another alternative embodiment, such as Figure 3 The diagram shown is a flowchart illustrating a size-driven conflict determination method provided in an embodiment of this application. This method performs precise detection of three types of potential conflicts before reconfiguration based on a local size constraint graph, in order to avoid invalid or contradictory size propagation.
[0075] Step S501: As Figure 3 As shown in (a), after adding the size constraints corresponding to the target design conditions, if there are two or more upstream size nodes with definite target values pointing to the same size node and the corresponding solution values are inconsistent, it is determined that there is a multi-source driving conflict.
[0076] In a local dimensional constraint diagram, multi-source drive conflict is characterized by the same downstream dimensional node simultaneously receiving two or more dimensional drive commands from different sources with incompatible values. The upstream dimensional node refers to the dimensional node that issues the drive action in the directed edge, located at the originating end of the arrow; it determines the value of the controlled downstream node. A defined target value means that the dimension is locked by a specific value or fixed design conditions, leaving no room for flexible adjustment. This type of conflict reflects the constraint contradiction in the equation system, i.e., a variable is forced to simultaneously equal multiple different values.
[0077] During the specific testing process, each dimension node in the local dimension constraint graph is traversed, and its in-degree, i.e., the number of directed edges pointing to that node, is counted. If the in-degree of a dimension node is greater than or equal to 2, it is further checked whether all upstream dimension nodes pointing to that node have definite target values. When the calculated solution values at the downstream nodes are not equal after the target values of these upstream nodes are calculated, it is determined that there is a multi-source driving conflict. For example, if one upstream node requires the downstream node size to be 100mm, while another upstream node requires it to be 120mm, this constitutes such a conflict. This situation usually occurs in scenarios where the old and new design conditions are not fully compatible. For example, holes in the original model driven by internal contour dimensions may be redefined by new external interface positions during modification, resulting in the hole size facing a dual and contradictory definition.
[0078] To avoid misinterpreting CAD numerical rounding errors as multi-source drive conflicts, the dimensions of each upstream target can first be converted into predicted dimension values for the same driven dimension node v using the geometric constraints of the corresponding directed edges. Then, the solution discrete quantities between different drive sources can be calculated. Specifically, this can be expressed as: ; In the formula, Pointing to the size node The One upstream size node; This represents the number of upstream nodes with a defined target value that are participating in the current conflict determination. For upstream size nodes Determine the target size value; (·) represents a directed edge. → The recorded geometric constraint transfer relationship is determined by the distance, coaxiality, angle, coincidence or other geometric constraints corresponding to the edge; Based solely on the first The size node calculated by the upstream driving source The predicted size value; For each upstream driving source at the size node The maximum predicted size difference generated at that location; For size nodes in-degree; For size nodes The consistency tolerance for the solution can be determined by taking the more stringent value between the corresponding dimensional design tolerance and the CAD numerical solution tolerance. For multi-source driven conflict markers, This indicates the existence of multi-source driver conflicts. This can be addressed by introducing... Conflict handling is triggered only when the difference in predicted values caused by different driving sources exceeds the allowable range of the project.
[0079] Step S502: As Figure 3 As shown in (b), when a directed path is formed after adding size constraints, starting from any size node and returning to that size node, a closed-loop driving conflict is determined to exist.
[0080] A closed-loop driven conflict refers to the existence of at least one closed path in the local dimension constraint graph, meaning that traveling along the driving direction indicated by the directed edge can start from a certain initial dimension node, pass through a series of intermediate nodes, and finally return to the initial node. The directed path represents the chain of dimension value transmission and the order of dependencies. The formation of a loop in the path means that there is a circular reference in the dimension driving chain.
[0081] Path traversal algorithms such as Depth-First Search (DFS) or Breadth-First Search (BFS) are used to trace paths through the local size constraint graph. During traversal, the visit history of the current path is recorded. Whenever a new node is attempted to be visited, it is checked whether the node already exists in the current path stack. If the node to be visited coincides with a node in the path stack, it proves that a directed cycle has been detected, starting from that node and returning to itself via subsequent nodes. This conflict indicates that the size constraint system is in an underdetermined or overdetermined state, physically manifested as the inability to determine the starting point or propagation direction of size adjustments. Since changes at any end will propagate along the cycle to itself, logical deadlock or numerical oscillations can occur, preventing the local size solver from obtaining a convergent unique solution.
[0082] Step S503: As Figure 3 As shown in (c), a path search is performed along the directed edge from the dimension node corresponding to the target design condition. When the search path reaches the dimension node associated with the functional topology anchor point and the target design condition requires the associated dimension to change, it is determined that there is an anchor point constraint conflict.
[0083] Anchor point constraint conflict refers to a newly introduced target design condition that attempts to alter the dimensions associated with functional topology anchor points that should remain unchanged, thereby compromising the core functionality of components or assembly compatibility. Dimensional nodes associated with functional topology anchor points are those that directly define the location of mounting interfaces, positioning reference relationships, or the geometry of critical functional structures; these nodes have been marked as protected objects during the influence domain determination process.
[0084] First, the starting dimension node corresponding to the target design condition is identified as the search source. A forward path search is performed along the directed edges in the local dimension constraint graph to simulate the propagation flow of dimension values. During the search, it is continuously determined whether the dimension node reached by the current path is associated with a functional topology anchor point attribute. If the path search reaches such a protected dimension node, and the change in the target design condition forces a change in the node's value (e.g., moving the positioning reference plane), then an anchor point constraint conflict is identified. This conflict identification directly warns that the current modification scheme may lead to the inability to install components or functional failure. It is necessary to avoid intrusion into the anchor point associated dimensions by adjusting the dimension chain in the non-anchor point area during the dimension reconfiguration stage.
[0085] The three conflict detection mechanisms described above together constitute the pre-judgment logic for size-driven relationship reconfiguration. By identifying multi-source driving conflicts, the system can pinpoint the source of the conflict; by identifying closed-loop driving conflicts, the system can discover unsolvable circular dependencies; and by identifying anchor point constraint conflicts, the system can prevent accidental damage to critical functional structures. This provides accurate decision-making basis and scope for subsequent steps such as removing low-level constraints, reversing driving directions, or segmenting size chains.
[0086] Example 7 In another embodiment, the process for locating directed edges that cause size-driven conflicts provided in this application includes locating the directed edges that cause size-driven conflicts, specifically including the following steps.
[0087] Step S601: Determine the node with the highest dimensional value and constraint level in the conflict path as the baseline driving node.
[0088] After detecting dimension-driven conflicts, to implement accurate dimensional relationship reconstruction, it is necessary to establish the core constraint basis in the conflict path, namely the reference driving node. This process involves traversing the conflict paths identified in the previous steps, performing attribute analysis on each dimension node involved in the path, focusing on whether the dimension node has a definite target value and its constraint level. The constraint level reflects the importance of the dimension in the functional structure of the component. For example, interface-locked dimensions directly limited by external assembly conditions usually have the highest level, while variable dimensions used for local structural adjustments have a lower level. By comparison and screening, the reference driving node is obtained. This node represents a strong constraint condition that must be prioritized and cannot be shaken in the current design scenario. When multiple dimension nodes have target values, the one with the highest constraint level is selected as the reference driving node, thereby ensuring that the subsequent reconstruction logic revolves around the most critical design intent. For example, in a path involving multi-source driving conflicts, if there is a hole pitch dimension directly locked by the customer installation interface and an internal structural contour dimension, and the two conflict, then the hole pitch dimension is determined as the reference driving node because it has a definite external target value and belongs to the high-level interface-locked dimension. This benchmark driver node provides a basis for determining which driving relationships need to be adjusted, and any driving relationship that contradicts its target constraints is considered a potential refactoring object.
[0089] Step S602: Search upstream dimension nodes in reverse along the direction pointing to the base driver node until a dimension node in a variable state with a constraint level lower than the base driver node is found, or a dimension node associated with a functional topology anchor point is found.
[0090] Starting from a defined baseline driving node, a reverse tracing search is performed to locate the nodes in the root chain of conflict that can be intervened upon. In the local dimensional constraint diagram, traversal is performed along the reverse direction of the directed edges (i.e., the input direction pointing to the baseline driving node), checking upstream dimensional nodes one by one. The termination condition for the reverse search is set to find nodes that meet specific characteristics: first, dimensional nodes in a variable state with constraint levels lower than the baseline driving node; these nodes typically correspond to free variables or low-priority structural dimensions that can be adjusted in the design, making them ideal candidates for absorbing conflict effects; second, dimensional nodes associated with functional topology anchors; these nodes relate to the core installation and positioning functions of components and are generally not allowed to be arbitrarily modified, reaching this point signifies that the boundary requiring protection has been reached. During the search process, the state (fixed or variable), constraint level, and connection relationship with the baseline driving node are recorded for each upstream dimensional node encountered along the way. For example, if the reference drive node is the mounting hole location, its upstream may be connected to the boss location size and the main body contour size. The reverse search will evaluate these upstream dimensions one by one. If the main body contour size is found to be in a variable state and at a low level, the search process will pause at that node and lock that node as the end of the reconstruction operation.
[0091] Step S603: Determine the directed edges in the reverse search path that are located between the reference driving node and the size node in the variable state, and whose driving direction conflicts with the target size constraint of the reference driving node, as driving edges to be reconstructed.
[0092] After completing the reverse search and identifying the upstream and downstream nodes of the target, conflict determination is performed on the directed edges within the search path to accurately filter out the driving relationships that need to be reconstructed. The determination of the driving edges to be reconstructed is based on the intersection of two dimensions: first, the positional dimension, which focuses only on the reverse search path between the base driving node and the variable-state size node determined in step S602, excluding irrelevant associations outside the path; second, the logical dimension, which checks whether the driving direction of the directed edge on this path has an essential conflict with the target size constraint of the base driving node. This conflict manifests as the base driving node being unable to reach its determined target value or being forced to deviate from its locked constraint state when the size change is transmitted according to the original driving direction. By combining the above two dimensions, specific directed edges that meet the conditions are marked as driving edges to be reconstructed. This process actually transforms the abstract size conflict problem into a specific graph theory edge set deletion or reversal operation. For example, in a chain where the hole position dimension is driven by the profile dimension, if the hole position dimension has been locked as the new reference driving node, while the profile dimension is in a variable state, the original directed edge between them (from the profile to the hole) is identified as a driving edge to be reconfigured because its driving direction conflicts with the new constraint of the locked hole position. Subsequent dimension reconfiguration steps will perform release operations or direction reversals on these driving edges to be reconfigured, thereby eliminating the conflict and establishing a new dimension driving relationship that meets the target design conditions.
[0093] Through the steps described above, utilizing the topology of the local dimension constraint graph, the specific driving relationship causing the conflict is automatically and accurately located after a dimension-driven conflict is detected. This method establishes an unshakeable benchmark based on the constraint level and locks in the adjustable range through a reverse tracing mechanism, ultimately separating the driving edge to be reconstructed from the complex dimension network. This location mechanism avoids blindly disturbing irrelevant dimension relationships when resolving conflicts, ensuring minimal intervention and targeted nature of dimension reconfiguration operations, and effectively supporting flexible adjustments to local dimension-driven relationships while maintaining the core functional topology unchanged.
[0094] Example 8 like Figure 4 As shown, in one optional implementation, a process for redetermining the driving relationship and dimensional propagation direction of corresponding dimensions is provided for embodiments of this application. The method further includes: Step S701: Delete the driving edge to be reconstructed and use the base driving node as the starting node for reconfiguration.
[0095] After identifying the directed edge causing the dimension-driven conflict and determining the baseline driving node, the driving edge to be reconfigured is first removed from the local dimension constraint graph. This operation cuts off the original dimension propagation logic that caused the conflict, freeing up logical space to establish a new driving relationship. Subsequently, the dimension node with a definite target value and the highest constraint level, i.e., the aforementioned determined baseline driving node, is set as the starting point for this dimension-driven relationship reconfiguration. This baseline driving node represents the insurmountable design constraints in this modification process, such as interface dimensions directly limited by external assembly conditions. Using it as the starting node for reconfiguration ensures that the subsequent reconfigured dimension chain will strictly extend outward from this core constraint.
[0096] Step S702: Following the reverse search path, new directed edges are sequentially established from the reference driving node to the dimension node in the variable state, opposite to the original driving direction, so that at least one driven dimension in the original dimension chain is converted into a driving dimension and at least one original driving dimension is converted into a driven dimension.
[0097] Starting with the reconfiguration start node as the logical starting point, new driving relationships are constructed downstream of the variable-state dimension nodes along the reverse search path determined in the above embodiments. This process is achieved by adding new directed edges to the local dimension constraint graph, and the direction of the new directed edges is opposite to the direction of the deleted original directed edges, which is essentially a logical reversal of the original dimension driving chain. For example, if dimension node A in the original model determines dimension node B as the driving dimension (i.e., there is a directed edge from A to B), and under the new design conditions, dimension node B is set as the base driving node, then a new directed edge from B to A is established. Through this reversal operation, the driven dimension B in the original dimension chain is converted into the new driving dimension, while the original driving dimension A is converted into a driven dimension, thereby changing the dimension propagation logic from adapting to the external interface to having the internal structure defined by the external interface, eliminating conflicts caused by bidirectional or multi-source driving.
[0098] Step S703: When the reconfiguration path reaches the dimension node associated with the functional topology anchor point or the dimension node in a fixed state, stop creating new directed edges so that the reconfigured dimension propagation path is limited to the local modification influence domain.
[0099] During the establishment of new directed edges, the extension range of the reconfiguration path is monitored in real time. When the path extends to a dimension node associated with a functional topology anchor point, or reaches a dimension node in a fixed state, the establishment of new directed edges is immediately stopped. Functional topology anchor points represent installation interfaces, positioning references, or critical functional structures that must remain unchanged, and the dimension nodes associated with them are the natural boundaries for propagation; while dimension nodes in a fixed state typically involve non-adjustable structural performance dimensions or externally locked dimensions. By terminating propagation at these boundary nodes, it is ensured that the reconfigured dimension propagation path is strictly limited to the previously determined local modification influence domain. This boundary control mechanism prevents dimensional changes from spreading to non-target areas, guarantees the topological stability of the model structure outside the functional topology anchor point, and achieves local driving logic reconfiguration without compromising the overall design intent.
[0100] In the aforementioned technical process, by deleting conflicting edges, reconstructing the reverse path starting from high-level constraints, and truncating boundaries using functional topology anchors, the complex dimensional chains that originally contained multi-source or closed-loop conflicts are reorganized into an ordered driving relationship that originates from the new design conditions, propagates unidirectionally, and is closed within the domain. This reconfiguration method not only resolves numerical conflicts but also logically empowers the new design conditions to dominate the local structure, providing unambiguous parameter calculation basis for subsequent local model regeneration.
[0101] Example 9 like Figure 4 As shown, in one optional embodiment, a process for allocating dimensional changes to variable-size nodes is provided for embodiments of this application. When the dimensional change caused by the target design conditions needs to be absorbed by two or more dimensional nodes in a variable state, the variable-size nodes are selected sequentially according to the constraint level from low to high. This step aims to establish the priority order of dimensional adjustments, ensuring that dimensional changes are absorbed first in areas with the least impact on the overall function of the component, thereby preserving the design intent of the critical structure to the greatest extent. First, all dimensional nodes in a variable state within the influence domain of the local modification are traversed, and their constraint level identifiers recorded in the dimensional constraint diagram or attribute table are retrieved. For example, according to the level system defined in the above embodiments, the constraint levels from high to low are interface locking dimensions, functional reference dimensions, structural performance dimensions, and variable dimensions used for local structural adjustments. The candidate dimensional nodes are sorted according to the constraint level values from small to large (i.e., constraint importance from low to high), and a list of nodes to be processed is generated. When there are multiple size nodes with the same minimum constraint level, they are further sorted according to the distance of each node's current size value from the boundary of the allowable change range. Nodes with greater absorption potential are selected as the currently selected variable size nodes to reduce the number of nodes involved in the adjustment and simplify the complexity of local reconstruction.
[0102] Step S801: For the currently selected variable size node, determine the amount of change that can be absorbed based on the difference between its current size value and the upper or lower limit of the allowable change range, and allocate the portion of the size change that does not exceed the amount of change that can be absorbed to the current variable size node.
[0103] For the currently selected i-th variable-size node, read its current size value di and the allowed variation range [Li, Ui]. Let the current unallocated size variation be ΔDr; when ΔDr > 0, the node absorbs the variation along the direction of increasing size; when ΔDr < 0, the node absorbs the variation along the direction of decreasing size. The absorbable variation Ci of the i-th node in the current variation direction can be expressed as: ; In the formula, Indicates the first The amount of change that a variable-size node can absorb in the current direction of change; and These represent the upper and lower limits of the allowable variation range for this size node, respectively; This indicates the current size value of the node. Indicates entering the first The size change still needs to be allocated when there are 1 node. Because lie in[ , [Inside, therefore] It is a non-negative value; when This indicates that the size change has been completely absorbed by the preceding node.
[0104] In determining Then, the change to be allocated and The comparison is performed, and the actual size change undertaken by the current node, the updated node size, and the remaining unallocated size change are calculated, specifically as follows: ; In the formula, Indicates the first The actual signed size change undertaken by each variable-size node; (·) is the sign function, used to maintain the direction of size increase or decrease; The updated size after the current node has completed its allocation; This represents the remaining size change after absorption by the current node. When | |≤ At this time, the current node can completely absorb the remaining change. =0; when | |> At that time, the current node can only absorb The corresponding maximum allowable change amount, and reaching the interval boundary corresponding to the direction of change.
[0105] Step S802: When the size change is greater than the absorbable change of the current variable size node, adjust the current variable size node to the boundary value of the corresponding allowable change range, and continue to distribute the remaining size change to the variable size node of the next constraint level until the size change is completely absorbed or there is no variable size node that satisfies the allowable change range.
[0106] When | |> At that time, the current variable-size node will be... Adjust to the boundary of the allowable change range corresponding to the direction of change: adjust to when the size increases. Adjust to when size decreases The current node only absorbs the maximum amount of change it can withstand. And will be by The remaining dimensional changes are then distributed to the next variable-size node. This process ensures that the update result for each size node always satisfies... ≤ ≤ To avoid exceeding the allowable range of engineering specifications for local structural dimensions in order to meet the target design conditions.
[0107] Based on the aforementioned constraint level sorting logic, the next variable-size node is automatically selected, and the variable amount to be allocated in the next round is taken from the output of the previous node. Repeat steps S801 and S802. When =0, or its absolute value is not greater than the dimensional solution tolerance of the CAD system, it is considered that the target dimension change has been completely absorbed and the allocation stops; if the condition is still met after all variable dimension nodes have been traversed, | If the dimensional solution tolerance is not met, it is determined that the current local modification influence domain cannot fully absorb the target dimensional change within the given allowable variation range. The process then returns to the dimensional-driven relationship reconfiguration step, where a new variable-size node, alternative propagation path, or driving edge direction is selected. The dimensional solution tolerance uses the numerical tolerance of the CAD parametric solver or a more stringent value from the relevant dimensional design tolerances.
[0108] Example 10 like Figure 5 As shown, in another optional embodiment, after the local regeneration is completed according to the target size driving relationship, the functional topology anchor points in the regenerated model are checked for topology inheritance. This is a process for checking the topology inheritance of functional topology anchor points provided in this application embodiment. This check process is used to ensure that the modified model retains the key design intent in the parent model.
[0109] Step S901: Extract candidate geometric entities from the regenerated model that are of the same geometric entity type as each functional topological anchor point in the parent model. Based on the orientation and positional relationship of the candidate geometric entities relative to the reference system and the topological adjacency relationship with adjacent geometric entities, match them item by item with the anchor point description data of the corresponding functional topological anchor points.
[0110] The regenerated model refers to the geometric state of the target component model obtained after performing a local feature regeneration operation. During local modification, although the functional topology anchor points are set to remain unchanged, the original geometric entity references in the parent model may be broken due to operations such as face merging, edge splitting, or ID (Identity Document) renumbering that the CAD (Computer-Aided Design) system may automatically perform when updating the geometric model. Therefore, the continuity of geometric IDs cannot be relied upon to determine whether anchor points are inherited; matching must be performed based on the essential attributes of the geometric entities.
[0111] The geometric topology of the regenerated model is traversed, and objects with the same geometric entity type as the functional topological anchor points defined in the parent model are extracted as candidate geometric entities. The criteria for determining the geometric entity type include: for surface entities, distinguishing between planes, cylindrical surfaces, spheres, cones, etc.; for edge entities, distinguishing between straight edges, arc edges, spline curve edges, etc. For example, if a functional topological anchor point in the parent model is a positioning cylindrical hole, then all cylindrical surfaces in the regenerated model are listed as candidate geometric entities.
[0112] The actual geometric attributes of the candidate geometric entities are compared item by item with the pre-established anchor point description data in the parent model. The anchor point description data records the key topological features of the functional topological anchors. The comparison process mainly includes the following three dimensions: First, the datum relationship dimension: Calculate the directional and positional relationships of candidate geometric entities relative to the global or local datum system in the regenerated model. Directional relationships are determined by the angle between the normal vector or axis vector and the datum vector, while positional relationships are determined by the distance or coordinate values after coordinate transformation. For example, check whether the axis of a candidate cylindrical surface is still perpendicular to the specified mounting datum surface, and whether the distance from the center point to that datum surface is within a preset tolerance range.
[0113] Second, the topological adjacency dimension: analyzing the connections between candidate geometric entities and their surrounding geometric entities. This includes which faces or edges the entity is adjacent to, the order of adjacency, and the type of adjacency (e.g., tangent, intersecting, coplanar, etc.). For example, if the original anchor point description data records that the plane is adjacent to the top surface of the boss and forms a rounded transition with the side surface, then the candidate geometric entities are searched for planes that satisfy the same adjacency relationship.
[0114] Third, the item-by-item matching logic: The directional, positional, and topological adjacency relationships obtained from the actual measurements are logically ANDed with the theoretical values or constraints recorded in the anchor point description data. Only when a candidate geometric entity conforms to the definition in the description data across all key dimensions is the entity considered to have passed the matching check. This multi-dimensional geometric attribute matching method overcomes the identification difficulties caused by changes in topological IDs due to local model reconstruction, ensuring the accuracy of the verification results.
[0115] To ensure that the above item-by-item matching has a quantifiable criterion for repeatable execution, for the functional topological anchor point 'a' in the parent model and the candidate geometric entity 'c' in the regenerated model, the orientation deviation, positional deviation, and topological adjacency consistency can be calculated respectively, and the inheritance matching determination can be formed accordingly. Specifically, this can be expressed as: ; In the formula, and These are the functional topology anchor points of the parent model. With candidate geometric entities The normalized unit direction vector; and These are the position vectors of the two relative to the same reference system; and These are directional deviation and positional deviation, respectively. and Anchor points With candidate entities The topological adjacency description set, each set element is composed of the adjacency entity type, adjacency relationship type and connection boundary geometry type, and does not use Face ID or Edge ID as set elements; For topological adjacency consistency, when and When it is an empty set, Defined as 1; and Anchor points With candidate entities The geometric entity type; and These are the allowable directional and positional deviations for the anchor point, respectively. This represents the lowest topological adjacency consistency of the anchor point. To inherit the matching tag, =1 indicates a candidate entity Satisfy the corresponding functional topology anchor points The conditions for inheritance. and Determined based on the corresponding assembly angle tolerance, positioning tolerance, and CAD geometric solution tolerance. The key adjacency relationships marked as essential to maintain in the anchor point description data are determined; all such adjacency relationships should be matched. For entities such as spheres that do not have a unique direction vector, the direction deviation condition can be omitted; for entities whose axes have the same functional significance in opposite directions, the absolute value of the dot product of the direction vectors is taken.
[0116] Step S902: The candidate geometric entities that simultaneously satisfy the geometric entity type, datum relationship, and topological adjacency relationship are determined as the inheriting entities of the corresponding functional topological anchor points.
[0117] Inherited entities can refer to geometric objects that are successfully identified and mapped back to the functional semantics of the parent model in the regenerated model. When a candidate geometric entity satisfies the geometric entity type, orientation / position deviation, and topological adjacency consistency metric criteria given in step S901, i.e., the corresponding inheritance matching tag M(a,c)=1, the candidate geometric entity is determined as the inherited entity of the functional topological anchor point a. This determination allows for minor geometric deviations within the assembly tolerance and CAD numerical solution tolerance range, but does not allow for the destruction of the anchor point's critical adjacency relationship or for deviations exceeding its functional allowable orientation or position range, thereby avoiding misjudging normal regeneration results caused only by changes in Face ID or Edge ID as topological inheritance failure.
[0118] This determination process essentially establishes a semantic-based mapping relationship, logically connecting entities that may have undergone physical ID changes or minor geometric adjustments back to the original design semantic system. By confirming the existence of inherited entities, it can be verified whether the local regeneration operation has successfully preserved the original functional interfaces and positioning benchmarks while meeting the new design conditions. For example, if the parent model defines a ring-shaped plane for sealing mating as a functional topology anchor point, and the regenerated model contains a ring-shaped plane whose spatial position, normal direction, and surrounding connection hole relationships all match the description data, then this ring-shaped plane is identified as an inherited entity. This means that the sealing function has been effectively preserved after the modification.
[0119] Step S903: When any functional topology anchor does not have a corresponding inherited entity, the size propagation path connected to the functional topology anchor is determined as a failure path, and the size-driven relationship reconfiguration step is returned to, and the size change amount of the variable size nodes in the failure path is reassigned.
[0120] After completing the matching and verification of all functional topology anchors, if any functional topology anchor is found to have no corresponding inheriting entity, this indicates that during the previous local regeneration process, the key structure represented by this anchor has undergone unacceptable changes, disappeared, or been severely deformed, resulting in the inability to meet the functional constraints defined by the anchor description data. In this case, the modification result is determined to be unqualified at the topology inheritance level, and the corresponding repair mechanism is initiated.
[0121] By tracing the relationship between the failure function's topological anchor point and the dimension-driven graph, all dimension propagation chains that directly or indirectly cause changes in the anchor point's geometry are identified as failure paths. A failure path typically includes several dimension nodes extending from the source feature to the anchor point and their driving relationships.
[0122] After identifying the failure path, a backtracking operation is performed, returning to the dimension-driven relationship reconfiguration step. At this point, instead of directly modifying the target design conditions, adjustments are made to the dimension constraint strategies along the failure path. Specifically, the dimension nodes involved in the failure path are re-analyzed, especially those in variable states with lower constraint levels. By adjusting the allocation of dimension changes for these variable dimension nodes—for example, reducing the change in a certain dimension, enabling alternative dimension absorption paths, or changing the order of dimension adjustments—an attempt is made to avoid destructive impacts on functional topology anchors. Through this iterative reconfiguration and verification loop, a dimension-driven solution that satisfies the new design conditions and fully inherits the critical functional topology can be automatically found, thus ensuring the reliability and robustness of the modified design.
[0123] Through the above verification and backtracking mechanism, the method provided in this application embodiment can automatically identify and repair the problem of damage to functional topology after partial modification, ensuring that the target component model strictly maintains the original assembly positioning and functional cooperation relationship while adapting to the new installation space or interface requirements.
[0124] The above description is merely a preferred embodiment of this application and is not intended to limit this application. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the protection scope of this application.
Claims
1. A method for component topology inheritance and size reconfiguration for local modification, characterized in that, include: Obtain the parent model and target design conditions of the component to be modified, and analyze the feature dependencies, topological adjacency relationships and dimension-driven relationships of the parent model; The functional topology anchor points are determined based on the installation interfaces, positioning references, and functional structures in the parent model, and the characteristics of the source of change are determined by comparing the target design conditions with the original design conditions. Starting from the source of change features, the influence propagation proceeds along the feature dependency and size-driven relationships, and terminates when the propagation reaches the preservation boundary corresponding to the functional topology anchor point, in order to determine the local modification influence domain. The dimensional constraint level is determined based on the functional constraint attributes of each dimension within the influence domain of the local modification, and the dimensional constraints corresponding to the target design conditions are added to the original dimensional driving relationship. When a size-driven conflict occurs after the size constraint is added, the size constraint with the higher constraint level is maintained, the original driving relationship of the lower constraint level size that conflicts with it is released, and the driving relationship and size propagation direction of the corresponding size are re-determined to obtain the target size driving relationship. According to the target size driving relationship, the features within the local modification influence domain are solved for size and locally regenerated, while keeping the model structure outside the local modification influence domain and the topological relationship corresponding to the functional topological anchor point unchanged, to obtain the target component model.
2. The method according to claim 1, characterized in that, The determination of the functional topology anchor points includes: Extract the geometric entities corresponding to the installation interface, positioning reference and functional structure from the parent model, and establish anchor point description data for each geometric entity. The anchor point description data includes at least the geometric entity type, the orientation and positional relationship relative to the reference system, the topological adjacency relationship with adjacent geometric entities and the association relationship with the dimension nodes. Geometric entities with the same functional identifier and whose anchor point description data are related to each other are grouped into a functional topology anchor point, so that the functional topology anchor point is associated with the corresponding functional structure in the parent model through the anchor point description data rather than the geometric entity number.
3. The method according to claim 2, characterized in that, The preservation boundary corresponding to the functional topology anchor point is determined in the following way: Using the geometric entities contained in the functional topology anchor points as initial entities, extract the geometric entities directly adjacent to the initial entities and the structural features associated with the initial entities through positioning dimensions or mating dimensions; The structural features used to generate the functional topology anchor point, restrict the position of the functional topology anchor point relative to the reference system, or maintain the connection relationship between the functional topology anchor point and adjacent structures are determined as boundary features, and the functional topology anchor point and the boundary features constitute the retention boundary.
4. The method according to claim 3, characterized in that, The determination of the influence domain of the local modification includes: Using the change source feature as the current affected feature, the next size node is determined along the size-driven relationship based on the change size associated with the current affected feature, and the structural feature corresponding to the next size node is determined as a candidate propagation feature. When the generation parameters of the candidate propagation feature are driven by the previous size node, or when the candidate propagation feature references the geometric entity generated by the previous affected feature, the candidate propagation feature is marked as the affected feature and continues to propagate. When the candidate propagation feature belongs to the functional topology anchor point or the boundary feature, or when further changing the candidate propagation feature will change the directional relationship, positional relationship or associated size corresponding to the functional topology anchor point, the propagation path pointing to the candidate propagation feature is truncated. The affected features and their associated dimensions covered by the untruncation propagation path are defined as the local modification influence domain.
5. The method according to any one of claims 1-4, characterized in that, A local dimension constraint graph is established for the local modification influence domain. The local dimension constraint graph uses dimensions as dimension nodes and the driving relationship between one dimension and another dimension as directed edges. Each dimension node records at least the current dimension value, allowed variation range, constraint level, and fixed or variable state of the corresponding dimension; each directed edge records at least the driving dimension, the driven dimension, and the geometric constraint relationship between the driving dimension and the driven dimension. The constraint levels, in descending order, include at least the interface locking dimensions directly defined by external assembly conditions, the functional reference dimensions defining positioning or mating relationships, the structural performance dimensions defining structural performance, and the variable dimensions used for local structural adjustments.
6. The method according to claim 5, characterized in that, The size-driven conflict is determined by the local size constraint diagram, including: After adding the size constraints corresponding to the target design conditions, if it is detected that there are two or more upstream size nodes with definite target values pointing to the same size node and the corresponding solution values are inconsistent, it is determined that there is a multi-source driving conflict. When a directed path is formed after adding the size constraint, starting from any size node and returning to that size node, a closed-loop driving conflict is determined to exist; A path search is performed along the directed edge from the dimension node corresponding to the target design condition. When the search path reaches the dimension node associated with the functional topology anchor point and the target design condition requires the associated dimension to change, an anchor point constraint conflict is determined to exist.
7. The method according to claim 6, characterized in that, When the size-driven conflict exists, the directed edge causing the size-driven conflict is located, including: The node with the highest constraint level and a defined target value in the conflict path is identified as the baseline driving node. Search upstream dimension nodes in reverse along the direction pointing to the reference driving node until a dimension node in a variable state with a constraint level lower than the reference driving node is found, or a dimension node associated with a functional topology anchor point is found. The directed edges in the reverse search path that are located between the reference driving node and the size node in the variable state, and whose driving direction conflicts with the target size constraint of the reference driving node, are identified as driving edges to be reconstructed.
8. The method according to claim 7, characterized in that, The process of redetermining the driving relationship and dimension propagation direction for the corresponding dimensions includes: Delete the driving edge to be reconstructed and use the baseline driving node as the reconfiguration starting node; According to the reverse search path, new directed edges opposite to the original driving direction are sequentially established from the reference driving node to the dimension node in the variable state, so that at least one driven dimension in the original dimension chain is converted into a driving dimension and at least one original driving dimension is converted into a driven dimension. When the reconfigured path reaches a dimension node associated with a functional topology anchor point or a dimension node in a fixed state, the creation of the new directed edge is stopped so that the reconfigured dimension propagation path is restricted to the local modification influence domain.
9. The method according to claim 8, characterized in that, When the dimensional change caused by the target design conditions needs to be absorbed by two or more dimensional nodes in a variable state, the variable dimensional nodes are selected in order of constraint level from low to high. For the currently selected variable-size node, the amount of change that can be absorbed is determined based on the difference between its current size value and the upper or lower limit of the allowable change range, and the portion of the size change that does not exceed the amount of change that can be absorbed is allocated to the current variable-size node. When the size change exceeds the absorbable change of the current variable size node, the current variable size node is adjusted to the boundary value of the corresponding allowable change range, and the remaining size change is further distributed to the variable size node of the next constraint level until the size change is completely absorbed or there is no variable size node that satisfies the allowable change range.
10. The method according to claim 9, characterized in that, After completing the local regeneration according to the target size-driven relationship, the functional topology anchors in the regenerated model are subjected to topology inheritance verification, including: Candidate geometric entities of the same geometric entity type as each functional topological anchor point in the parent model are extracted from the regenerated model. Based on the orientation and positional relationship of the candidate geometric entities relative to the reference system and the topological adjacency relationship with adjacent geometric entities, they are matched item by item with the anchor point description data of the corresponding functional topological anchor points. Candidate geometric entities that simultaneously satisfy geometric entity type, baseline relationship, and topological adjacency relationship are determined as the inheriting entities of the corresponding functional topological anchor points; When any functional topology anchor does not have a corresponding inheriting entity, the size propagation path connected to that functional topology anchor is determined as a failure path, and the size-driven relationship reconfiguration step is returned to reallocate the size change amount of the variable size nodes in the failure path.
