CN-121997018-A - Method and system for intelligently identifying boundary characteristics of closed cavity based on geometric topology information

CN121997018ACN 121997018 ACN121997018 ACN 121997018ACN-121997018-A

Abstract

The invention discloses a closed cavity boundary characteristic intelligent recognition method and system based on geometric topology information, and the method comprises the steps of constructing a full-flow automatic method of boundary part recognition, inner and outer surface recognition, conducting surface recognition and notch recognition, screening boundary parts through coordinate transformation and distance detection, respectively completing inner and outer surface recognition through a B-Rep space scattering recognition algorithm and a graph theory search algorithm, realizing conducting surface accurate recognition based on contact area detection and curvature verification, and finally completing notch recognition through surface screening, parallelism verification and non-contact area detection. The invention solves the problems of low efficiency and easy omission of traditional manual detection, realizes the automation and high-precision identification of geometric features through a standardized algorithm, is suitable for the structural safety and tightness verification of the closed cavity in the fields of aerospace and the like, and can greatly improve the detection efficiency and accuracy.

Inventors

HUANG SHAOHUA
HE JIE
GUO YU
WANG FALIN
WU TAO
PU JUN
CHEN ZHIYI

Assignees

南京航空航天大学

Dates

Publication Date: 20260508
Application Date: 20251229

Claims (10)

1. The intelligent recognition method for the boundary characteristics of the closed cavity based on the geometric topology information is characterized by comprising the following steps: S1, boundary part identification, namely acquiring an assembly coordinate system of an assembly body of a closed cavity and a modeling coordinate system of an imaginary cavity, and converting geometric data of all parts in the assembly body from the assembly coordinate system to the modeling coordinate system; S2, performing inner and outer surface recognition on the boundary part, namely sampling the ideal cavity surface row by row to obtain discrete sampling points when the inner surface is recognized, and then performing ray emission and contact detection operation; S3, identifying a local contact surface between the boundary part and the virtual cavity, verifying whether the local contact surface meets constraint, and marking the local contact surface conforming to the constraint as a conduction surface; and S4, notch identification, namely screening two boundary parts with contact relation, detecting whether the two boundary parts have surface pairs which are not contacted but are in a contact range, verifying whether the surface pairs meet the constraint of alignment and parallelism, and marking the surface pairs conforming to the constraint as notches.
2. The intelligent recognition method for the boundary characteristics of the closed cavity based on the geometric topological information, which is disclosed in claim 1, is characterized in that the distance between the geometric data of the part and the geometric data of the virtual cavity in the step S1 comprises a point-point distance, a point-line distance, a point-surface distance, a line-line distance, a line-surface distance and a surface-surface distance, wherein a bounding box pre-detection mode is adopted in the calculation process, and non-candidate geometric element pairs are eliminated so as to improve the distance calculation efficiency.
3. The intelligent recognition method of the boundary characteristics of the closed cavity based on the geometric topology information according to claim 1, wherein the inner surface recognition implementation process in the step S2 is as follows: calling CATBRepAccess an interface to obtain a B-Rep model of the virtual cavity, sampling the surface of the ideal cavity line by line along the grid array, and generating a uniformly distributed discrete sampling point set Each sampling point comprises three-dimensional coordinates and normal vector information of the surface of the cavity; At each sampling point As origin, emitting fixed length rays along the direction opposite to the normal direction of the ideal cavity, i.e. pointing to the inside of the assembly Controlling the radiation emission through CATRay interfaces, calling CATIntersection interfaces to detect the contact state of the radiation and the triangular mesh surface patch of the boundary part in real time: If ray And a certain triangular mesh surface patch The contact is generated, and the included angle between the ray at the contact point and the normal vector of the surface patch is smaller than a preset value, the surface patch is marked Is an inner surface; Traversing all sampling points and rays to form an inner surface patch set 。
4. The intelligent recognition method of the boundary characteristics of the closed cavity based on the geometric topology information according to claim 1, wherein the external surface recognition implementation process in the step S2 is as follows: Regarding all triangular mesh patches of the boundary part as undirected graph Is the vertex of (2) If two face sheets have adjacent relation, an edge is established between the corresponding vertexes Invoking CATTopologicalQuery interfaces to analyze topological association relations among the patches, so as to ensure the accuracy of the adjacency graphs; inner surface patch set The patch in (a) is used as a starting vertex, a DFS algorithm is started to recursively search adjacent patches, and each candidate adjacent patch is subjected to Three constraint conditions are verified: and (3) dough sheet interval constraint: The minimum distance from the initial inner surface patch is less than or equal to 0.1mm; normal parallel constraint: An included angle between the normal vector of the (c) and the normal vector of the initial inner surface patch is less than or equal to 5 degrees; Same part constraint: The ID of the part is consistent with the ID of the part of the initial inner surface patch; Candidate patches satisfying three constraints simultaneously Marked as outer surface, finally forming an outer surface dough sheet set 。
5. The intelligent recognition method of the boundary characteristics of the closed cavity based on the geometric topology information according to claim 1, wherein the implementation process of the step S3 is as follows: Contact surface positioning based on an inner surface set of boundary parts Screening out a 'local contact area' from the surface of the virtual cavity B-Rep model, namely, an area where the inner surface patches overlap with the surface of the cavity or the distance between the inner surface patches is smaller than or equal to a contact threshold value, and taking the inner surface patches corresponding to the area as a conduction surface candidate patch set ; Curvature constraint verification, calling CATCurvature interface to calculate candidate patch set The curvature value of each dough sheet is compared with a preset curvature range: if the curvature value of the candidate dough piece is within the preset range and the dough piece is generated by the contact of the boundary part and the virtual cavity, marking the dough piece as a conducting surface, and finally forming a conducting surface set And providing a core detection object for cavity tightness verification.
6. The intelligent recognition method of the boundary characteristics of the closed cavity based on the geometric topology information according to claim 1, wherein the implementation process of the step S4 is as follows: Screening out part pairs with contact relation from the boundary part set, namely, the surface distance between two boundary parts is less than or equal to a contact judgment threshold value, or partial surface overlapping exists, and marking the part pairs as candidate part pairs ; For each candidate part Calling CATEdge interface to extract boundary curves of the two, and analyzing whether there are surface pairs which are not contacted but are in contact range or not Extraction surface in the inner/outer surface of (a) From the slave Extraction of dough sheets from the inner or outer surface of (a) If (if) And (3) with Between 0.05 and 0.5mm (in the contact range but without actual contact), will As a gap candidate surface pair; pair of candidate surfaces Two key constraints are verified: Position alignment constraints: geometric centroid of (2) The deviation between the geometric centroid connecting line and a preset reference direction is less than or equal to 3 degrees; Parallel constraint: Normal vector of (2) Included angle of normal vector of (2) less than or equal to 2 degrees; candidate surface pairs meeting both constraints The mark is a notch, and the position coordinates, the size and the part pair information of the notch are recorded.
7. A closed cavity boundary feature intelligent recognition system based on geometric topology information by adopting the method as set forth in any one of claims 1 to 6, comprising: The data preprocessing module is used for acquiring an assembly coordinate system of the closed cavity assembly and a modeling coordinate system of the virtual cavity, realizing the unification of the assembly coordinate system and the modeling coordinate system, and simultaneously completing the extraction and conversion of geometric data of all parts in the assembly; The boundary part identification module is used for calling CATMathTransformation interfaces of CATIA CAA to realize coordinate conversion of part geometric data, calling CATGeoFactory interfaces to extract point, line and surface geometric elements of the part, calculating the distance between the part and the virtual cavity based on Euclidean distance algorithm and a projection method, and screening out the boundary part by combining with pre-detection of a bounding box; The inner and outer surface recognition module is used for calling CATBRepAccess interfaces to obtain an ideal cavity BRep model, completing inner surface recognition through a BRep space scattering algorithm, constructing a patch adjacency graph based on CATTopologicalQuery interfaces, and completing outer surface recognition through a depth-first search algorithm; The conducting surface identification module calls CATCurvature an interface to calculate the curvature of the contact surface, verifies whether the contact surface meets the constraint that the contact is generated and the curvature is in a preset range, and marks the surface conforming to the constraint as a conducting surface; and a gap identification module, namely calling CATEdge interfaces to extract part boundary curves, verifying the constraint that the two parts are contacted and the surfaces are not contacted but are aligned in parallel, and marking the surface pair conforming to the constraint as a gap.
8. The intelligent recognition system for boundary features of closed cavities based on geometric topological information according to claim 1, wherein the system further comprises a user interaction module integrated on a CATIA interface, comprising a dedicated workbench and a command menu, supporting the setting of distance threshold and curvature range parameters, highlighting recognition results and deriving data.
9. The intelligent recognition system for boundary features of a closed cavity based on geometric topology information of claim 8, wherein the derived data comprises part ID, surface type and recognition state.
10. The intelligent recognition system for the boundary characteristics of the closed cavity based on the geometric topology information of claim 8, wherein the derived data supports multiple formats including Excel and TXT, so as to facilitate the subsequent analysis and archiving of recognition results by engineers.

Description

Method and system for intelligently identifying boundary characteristics of closed cavity based on geometric topology information Technical Field The invention belongs to the computer aided design and manufacturing technology, and particularly relates to a closed cavity boundary characteristic intelligent recognition technology based on geometric topology information. Background In the aerospace field, the tightness and structural integrity of the structures of airtight cavities such as an aircraft oil tank are directly related to the operation safety of equipment, medium leakage can be caused by abnormal contact of a conducting surface, stress concentration and fatigue damage are easily induced by a boundary part notch, and the service reliability of the whole aircraft is further threatened. The cavity assembly is formed by a plurality of boundary parts in a cooperative manner, the traditional detection mode depends on engineers to manually check geometric features in CATIA (ComputerAided Tri-Dimensional Interactive Application), the efficiency is low (the detection of a single cavity needs to consume 2-3 working days), the detection omission and the misjudgment are easily caused by the artificial subjective judgment deviation, and the severe requirements of modern aviation manufacturing on the detection precision and efficiency are difficult to meet. The prior art has three significant bottlenecks, namely, the CAD (Computer AIDED DESIGN) original function lacks batch processing capability of geometric data of the multi-boundary part, the unified difficulty of an assembly coordinate system and a cavity modeling coordinate system is high, the reliability of core judging links such as distance detection and the like is directly influenced, secondly, the inner and outer surface identification relies on engineer experience subjective judgment, quantitative judging standards are lacking, identification confusion is easy to occur when facing a complex curved surface structure, thirdly, the curvature matching verification of a conducting surface and the accurate identification of a notch contour do not have special automatic tool support, and particularly the detection coverage of a hidden notch in the assembly body is incomplete, and a significant omission risk exists. Therefore, a full-flow automatic closed cavity structure detection scheme capable of fundamentally solving the efficiency bottleneck and the accuracy defect of the traditional manual operation and greatly improving the detection efficiency of the closed cavity is needed. Disclosure of Invention The invention aims to provide an intelligent recognition method and an intelligent recognition system for boundary characteristics of a closed cavity based on geometric topological information, which solve the problems of low efficiency and difficult error control in the traditional manual detection mode, realize full-flow automatic recognition of boundary parts, boundary surfaces and gaps of the closed cavity by fusing an interference detection technology, geometric topological characteristic rules, CAA secondary development tools and a regularized characteristic matching algorithm, greatly improve detection precision and execution efficiency, and can be accurately adapted to complex assembly scene requirements of few samples and high-precision guidance in the field of aviation manufacturing. The intelligent recognition method for the boundary characteristics of the closed cavity based on the geometric topology information comprises the following steps: S1, boundary part identification, namely acquiring an assembly coordinate system of an assembly body of a closed cavity and a modeling coordinate system of an imaginary cavity, and converting geometric data of all parts in the assembly body from the assembly coordinate system to the modeling coordinate system; S2, performing inner and outer surface recognition on the boundary part, namely sampling the ideal cavity surface row by row to obtain discrete sampling points when the inner surface is recognized, and then performing ray emission and contact detection operation; S3, identifying a local contact surface between the boundary part and the virtual cavity, verifying whether the local contact surface meets constraint, and marking the local contact surface conforming to the constraint as a conduction surface; and S4, notch identification, namely screening two boundary parts with contact relation, detecting whether the two boundary parts have surface pairs which are not contacted but are in a contact range, verifying whether the surface pairs meet the constraint of alignment and parallelism, and marking the surface pairs conforming to the constraint as notches. Further, the distance between the geometric data of the part and the geometric data of the imaginary cavity in the step S1 comprises a point-point distance, a point-line distance, a point-plane distance, a line-line distance, a line-plane distance and