Boundary Representation Models Research Articles

Adaptive recognition of machining features in sheet metal parts based on a graph class-incremental learning strategy

The integration of computer-aided design (CAD), computer-aided process planning (CAPP), and computer-aided manufacturing (CAM) systems is significantly enhanced by employing deep learning-based automatic feature recognition (AFR) methods. These methods outperform traditional, rule-based approaches, particularly in handling the complexities of intersecting features. However, existing deep learning-based AFR methods face two major challenges. The initial challenge stems from the frequent utilization of voxelized or point-cloud representations of CAD models, resulting in the unfortunate loss of valuable geometric and topological information inherent in original Boundary representation (B-Rep) models. The second challenge involves the limitation of supervised deep learning methods in identifying machining features that are not present in the predefined dataset. This constraint renders them suboptimal for the continually evolving datasets of real industrial scenarios. To address the first challenge, this study introduces a graph-structured language, Multidimensional Attributed Face-Edge Graph (maFEG), crafted to encapsulate the intricate geometric and topological details of CAD models. Furthermore, a graph neural network, Sheet-metalNet, is proposed for the efficient learning and interpretation of maFEGs. To tackle the second challenge, a three-component incremental learning strategy is proposed: an initial phase of pre-training and fine-tuning, a prototype sampling-based replay, and a stage employing knowledge distillation for parameter regularization. The effectiveness of Sheet-metalNet and its complementary incremental learning strategy is evaluated using the open-source MFCAD++ dataset and the newly created SMCAD dataset. Experimental results show that Sheet-metalNet surpasses state-of-the-art AFR methods in machining feature recognition accuracy. Moreover, Sheet-metalNet demonstrates adaptability to dynamic dataset changes, maintaining high performance when encountering newly introduced features, thanks to its innovative incremental learning strategy.

BrepMFR: Enhancing machining feature recognition in B-rep models through deep learning and domain adaptation

Feature Recognition (FR) plays a crucial role in modern digital manufacturing, serving as a key technology for integrating Computer-Aided Design (CAD), Computer-Aided Process Planning (CAPP) and Computer-Aided Manufacturing (CAM) systems. The emergence of deep learning methods in recent years offers a new approach to address challenges in recognizing highly intersecting features with complex geometric shapes. However, due to the high cost of labeling real CAD models, neural networks are usually trained on computer-synthesized datasets, resulting in noticeable performance degradation when applied to real-world CAD models. Therefore, we propose a novel deep learning network, BrepMFR, designed for Machining Feature Recognition (MFR) from Boundary Representation (B-rep) models. We transform the original B-rep model into a graph representation as network-friendly input, incorporating local geometric shape and global topological relationships. Leveraging a graph neural network based on Transformer architecture and graph attention mechanism, we extract the feature representation of high-level semantic information to achieve machining feature recognition. Additionally, employing a two-step training strategy under a transfer learning framework, we enhance BrepMFR's generalization ability by adapting synthetic training data to real CAD data. Furthermore, we establish a large-scale synthetic CAD model dataset inclusive of 24 typical machining features, showcasing diversity in geometry that closely mirrors real-world mechanical engineering scenarios. Extensive experiments across various datasets demonstrate that BrepMFR achieves state-of-the-art machining feature recognition accuracy and performs effectively on CAD models of real-world mechanical parts.

BRepGAT: Graph neural network to segment machining feature faces in a B-rep model

Abstract In recent years, there have been many studies using artificial intelligence to recognize machining features in three-dimensional models in the computer-aided design (CAD)/computer-aided manufacturing field. Most of these studies converted the original CAD data into images, point clouds, or voxels for recognition. This led to information loss during the conversion process, resulting in decreased recognition accuracy. In this paper, we propose a novel deep learning model called the boundary representation graph attention network (BRepGAT) to segment faces in an original boundary representation (B-rep) model containing machining features. We define descriptors that represent information about the faces and edges of the B-rep model from the perspective of feature recognition. These descriptors are extracted from the B-rep model and transformed into homogeneous graph data, which are then passed to graph networks. BRepGAT recognizes machining features from the graph data input. Our experimental results using the MFCAD18++ dataset showed that BRepGAT achieved state-of-the-art recognition accuracy (99.1%). Furthermore, BRepGAT showed relatively robust performance on other datasets besides MFCAD18++.

AAGNet: A graph neural network towards multi-task machining feature recognition

Machining feature recognition (MFR) is an essential step in computer-aided process planning (CAPP) that infers manufacturing semantics from the geometric entities in CAD models. Traditional rule-based MFR methods struggle to handle intersecting features due to the complexity of representing their variable topological structures. This motivates the development of deep-learning-based methods, which can learn from data and overcome the limitations of rule-based methods. However, some existing deep learning methods compromise geometric and topological information when using certain representations such as voxel or point cloud. To address these challenges, we propose a novel graph neural network, named AAGNet, for automatic feature recognition using a geometric Attributed Adjacency Graph (gAAG) representation that preserves topological, geometric, and extended attributes from neutral boundary representation (B-Rep) models. Furthermore, some existing methods (such as UV-Net, Hierarchical CADNet) lack the capability of machining feature instance segmentation, which is a sub-task of feature recognition that requires the network to identify different machining features and the B-Rep face entities that constitute them, and it is a crucial task for subsequent process planning. AAGNet is designed as a multi-task network that can perform semantic segmentation, instance segmentation, and bottom face segmentation simultaneously for recognizing machining features, the faces associated with those features, and their bottom faces. The AAGNet is evaluated on various open-source datasets, including MFCAD, MFCAD++, and the newly introduced MFInstSeg dataset with over 60,000 STEP files and machining feature instance labels. The experimental results demonstrate that AAGNet outperforms other state-of-the-art methods in terms of accuracy and complexity, showing its potential as a flexible solution for MFR in CAPP.

Isogeometric analysis of 3D solids in boundary representation for problems in nonlinear solid mechanics and structural dynamics

AbstractThis contribution presents an isogeometric approach in three dimensions for nonlinear problems in solid mechanics and structural dynamics. The proposed approach aligns with the boundary representation modeling technique in CAD. Isogeometric analysis is combined with the parameterization of the scaled boundary finite element method. In this way, the boundary description of 3D solids in CAD can be directly utilized for the analysis in an isogeometric framework. The approximation of the solution on the boundary is based on bi‐variate NURBS. We also approximate the interior of the solid with uni‐variate B‐splines to facilitate the solution of nonlinear problems. The nonlinear case is extended to three dimensions in this present work. The method is further extended to dynamic problems. The mass and damping matrices are derived using the same basis functions. The solution of the entire solid is obtained with the Galerkin method. We study several nonlinear and dynamic problems with simple geometries and arbitrary number of boundaries. Moreover, a fiber reinforced composite serves as demonstration for complex shapes.

Machining feature recognition based on deep neural networks to support tight integration with 3D CAD systems

Recently, studies applying deep learning technology to recognize the machining feature of three-dimensional (3D) computer-aided design (CAD) models are increasing. Since the direct utilization of boundary representation (B-rep) models as input data for neural networks in terms of data structure is difficult, B-rep models are generally converted into a voxel, mesh, or point cloud model and used as inputs for neural networks for the application of 3D models to deep learning. However, the model’s resolution decreases during the format conversion of 3D models, causing the loss of some features or difficulties in identifying areas of the converted model corresponding to a specific face of the B-rep model. To solve these problems, this study proposes a method enabling tight integration of a 3D CAD system with a deep neural network using feature descriptors as inputs to neural networks for recognizing machining features. Feature descriptor denotes an explicit representation of the main property items of a face. We constructed 2236 data to train and evaluate the deep neural network. Of these, 1430 were used for training the deep neural network, and 358 were used for validation. And 448 were used to evaluate the performance of the trained deep neural network. In addition, we conducted an experiment to recognize a total of 17 types (16 types of machining features and a non-feature) from the B-rep model, and the types for all 75 test cases were successfully recognized.

A robust direct modeling method for quadric B-rep models based on geometry–topology inconsistency tracking

Boundary representation (B-rep) model editing plays an essential role in computer-aided design, and has motivated the very recent direct modeling paradigm, which features intuitive push–pull manipulation of the model geometry. In mechanical design, a substantial part of B-rep models being used are quadric models (composed of linear and quadric surfaces). However, push-pulling such models is not trivial due to the possible smooth face–face connections in the models. The major issue is that, during push–pull moves, it is often desirable to preserve these connections for functional, manufacturing, or aesthetic reasons, but this could cause complex inconsistencies between the geometry and topology in the model and lead to robustness issues in updating the model. The challenge lies in effectiveness towards detecting the instants when geometry–topology inconsistencies occur during push–pull moves. This paper proposes a novel reverse detection method to solve the challenge and then, based on it, presents a robust method for push–pull direct modeling while preserving smooth connections. Case studies and comparisons have been conducted to demonstrate the effectiveness of the method.

Digital Plan of Brickwork Layout for Robotic Bricklaying Technology

The trend of using modern technologies in the construction industry has been growing stronger recently, particularly in the fields of additive construction or robotic bricklaying. Therefore, specifically for the purpose of robotic bricklaying, we created a digital layout plan for robotic construction works. This article presents a universal methodology for creating a bricklaying plan for various variations of wall building systems. The method is based on the conversion of drawings from the BIM (Building Information Model) environment to the BREP (Boundary Representation) model through use of the IFC (Industry Foundation Classes) format, which simultaneously divides object models into layers and connects discontinuous wall axes by means of an orthogonal arrangement and inserting details into critical structural points. Among other aspects, the developed algorithm proposes the optimal placement of the robotic system inside objects under construction, in order to minimize the distance of the robot’s movement and to reduce its electricity consumption. Digital layout plans created in this way are expected to serve as a stepping stone for robotic bricklaying.

Multiobjective evolutionary optimization for feature-based simplification of 3D boundary representation models

Boundary representation (B-rep) is one of the most common techniques for representing three-dimensional (3D) shapes. In this method, shapes are described mathematically based on the vertices, edges, and faces present. B-rep models are generated using computer-aided design systems and are employed for various purposes in downstream applications, including computer-aided engineering, manufacturing, and inspection. Therefore, the level of detail of the 3D model should be adjusted based on the purpose of use. In this study, we performed multiobjective optimization for the feature-based simplification of B-rep models. The two objectives of the general model simplification process are to minimize the data size and limit the differences in appearance with respect to the original model. To simultaneously meet these objectives, model simplification was performed to minimize both the difference in the volume and the number of faces with respect to those of the original model. For optimization, we used the simple genetic algorithm (SGA) and the non-dominated sorting genetic algorithm II (NSGA-II). The experimental results confirmed that solution set obtained using NSGA-II was of higher quality as compared with those for SGA and the conventional method.

An approach to automatic boundary segmentation of solid models using virtual topology: toward reconstruction of design features

Abstract Boundary segmentation of solid models is the geometric foundation to reconstruct design features. In this paper, based on the shape evolution analysis for the feature-based modeling process, a novel approach to the automatic boundary segmentation of solid models for reconstructing design features is proposed. The presented approach simulates the designer’s decomposing thinking on how to decompose an existing boundary representation model into a set of design features. First, the modeling traces of design features are formally represented as a set of feature vertex adjacent graphs that use low-dimensional vertex entities and their connection relations. Then, a set of Boolean segmentation loops is searched and extracted from the constructed feature vertex adjacent graphs, which segment the boundary of a solid model into a set of regions with different design feature semantics. In the search process, virtual topology operations are employed to simulate the topological changes resulting from Boolean operations in feature modeling processes. In addition, to realize effective search, search strategies and search algorithms are presented. The segmentation experiments and case study show that the presented approach is feasible and effective for the boundary segmentation of medium-level complex part models. The presented approach lays the foundation for the later reconstruction of design features.

Patch coupling in isogeometric analysis of solids in boundary representation using a mortar approach

SummaryThis contribution is concerned with a coupling approach for nonconforming NURBS patches in the framework of an isogeometric formulation for solids in boundary representation. The boundary representation modeling technique in CAD is the starting point of this approach. We parameterize the solid according to the scaled boundary finite element method and employ NURBS basis functions for the approximation of the solution. Therefore, solid surfaces consist of several sections, which can be regarded as patches and discretized independently. The main objective of this study is to derive an approach for the connection of independent sections in order to allow for local refinement and thus an accurate and efficient discretization of the computational domain. Nonconforming sections are coupled with a mortar approach within a master‐slave framework. The coupling of adjacent sections ensures the equality of mutual deformations along the interface in a weak sense and is enforced by constraining the NURBS basis functions on the interface. We apply this approach to nonlinear problems in two dimensions and compare the results with conforming discretizations.

Development of a nonhuman primate computational phantom for radiation dosimetry.

The nonhuman primate (NHP) is an important animal model for evaluating the response of the human body to radiation exposure owing to similarities between its organ structure, genome, life span, and metabolism. However, there is a lack of radiation dosimetry estimations for NHPs. The aim of this work is to construct a computational phantom of NHPs and estimate absorbed fractions and specific absorbed fractions for internal radiation dosimetry. A female rhesus monkey was frozen and sectioned using a cryomacrotome. The transaxial sectioned images were imported into Adobe Photoshop for segmentation of internal organs. A total of 31 organs/tissues were identified and labeled. The segmented voxel phantom was then converted to a boundary representation (BREP) phantom to enable easy alteration of the phantom to mimic monkeys of different stature. The BREP model was then voxelized and imported into the MCNPX Monte Carlo radiation transport code for electron and photon dosimetry calculations. To estimate the appropriateness of using human phantoms as surrogate models for NHPs, absorbed fractions (AFs) and specific absorbed fractions (SAFs) of monoenergetic electrons and photons were calculated and compared with the ICRP reference newborn female phantom and a 1-yr-old female phantom. Considerable differences were observed for both self-absorbed and cross-absorbed doses for some organs between the NHP phantom and newborn phantom. For example, the ratios of the self-absorbed SAF(stomach wall) Monkey to SAF(stomach wall) Newborn ranged from 0.06 at 10keV to 0.29 at 3MeV for photons while the corresponding ratios to cross-absorbed SAF(liver→kidney) Monkey to SAF(liver→kidney) Newborn ranged from 0.78 at 50keV to 5.78 at 10keV for photons. Conversely, values of self-absorbed SAF were much higher (ratios of 2.39-4.19) for the brain and much lower for the uterus (0.51-0.61) in the monkey model compared to the newborn phantom. These dose differences can be attributed to the disparities between organ masses, shapes, and positions between the two phantoms. The developed NHP model can be exploited for the assessment of radiation dose to NHPs in preclinical radiation dosimetry studies and radiation therapy research.

A physically and geometrically nonlinear formulation for isogeometric analysis of solids in boundary representation

AbstractThis contribution concerns a physically and geometrically nonlinear formulation for isogeometric analysis of solids in boundary representation. The parametrization of the proposed approach is inspired by the scaled boundary finite element method, where the geometry of the boundary is sufficient to define the entire solid surface. This suits perfectly the boundary representation modeling technique, which is commonly employed for the design of solids in CAD. In order to provide a boundary‐oriented approach, the solid surface is parametrized by a radial scaling parameter emanating from a scaling center and a parameter in the circumferential direction of the boundary. According to the idea of isogeometric analysis, NURBS basis functions define the geometry of the boundary and approximate the solution on the boundary. We also employ NURBS to approximate the solution in the scaling direction in the interior of the domain. This enables a straightforward treatment of nonlinear problems while preserving the exact geometry of the boundary. The approximation in scaling direction is in principle flexible and requires for coupling of adjacent patches inside the domain. The Galerkin method is employed for the solution in both parametric directions and a linearization is derived which is used within an iterative Newton‐Raphson scheme. We consider nonlinear problems in solid mechanics including elasto‐plastic material behavior and large deformations as well as complex geometries commonly employed in engineering applications. In this paper we study a numerical example of a complex geometry in order to evaluate the performance of the proposed approach.

Research of Visual Conversion Technology for Reactor Core Monte Carlo Model Based on CSG

In order to improve the efficiency of 3D visualization analysis of the core Monte Carlo fine transport calculation model (MC calculation model), on the basis of fully investigating the current development status of the MC calculation model 3D visualization tools, and by analyzing the characteristics of different types of geometric models and the description methods of constructive solid geometry (CSG) models, this paper proposed an automatic 3D visualization converting method for the core fine MC calculation model, and conducted a deep study on the key technologies such as the conversion from CSG model to boundary representation (BREP) model, the mesh discretization of BREP model and the optimization of conversion efficiency for CSG model. The test results show that the proposed method can meet the engineering requirements in terms of function and efficiency.

Evaluation of Topological Consistency in CityGML

Boundary representation models are data models that represent the topology of a building or city model. This leads to an issue in combination with geometry, as the geometric model necessarily has an underlying topology. In order to allow topological queries to rely on the incidence graph only, a new notion of topological consistency is introduced that captures possible topological differences between the incidence graph and the topology coming from geometry. Intersection matrices then describe possible types of topological consistency and inconsistency. As an application, it is examined which matrices can occur as intersection matrices, and how matrices from topologically consistent data look. The analysis of CityGML data sets stored in a spatial database system then shows that many real-world data sets contain many topologically inconsistent pairs of polygons. It was observed that even if data satisfy the val3dity test, they can still be topologically inconsistent. On the other hand, it is shown that the ISO 19107 standard is equivalent to our notion of topological consistency. In the case when the intersection is a point, topological inconsistency occurs because a vertex lies on a line segment. However, the most frequent topological inconsistencies seem to arise when the intersection of two polygons is a line segment. Consequently, topological queries in present CityGML data cannot rely on the incidence graph only, but must always make costly geometric computations if correct results are to be expected.

Slicing and support structure generation for 3D printing directly on B-rep models

Traditional 3D printing is based on stereolithography or standard tessellation language models, which contain many redundant data and have low precision. This paper proposes a slicing and support structure generation algorithm for 3D printing directly on boundary representation (B-rep) models. First, surface slicing is performed by efficiently computing the intersection curves between the faces of the B-rep models and each slicing plane. Then, the normals of the B-rep models are used to detect where the support structures should be located and the support structures are generated. Experimental results show the efficiency and stability of our algorithm.

Variational B-rep model analysis for direct modeling using geometric perturbation

Abstract The very recent CAD paradigm of direct modeling gives rise to the need of processing 3D geometric constraint systems defined on boundary representation (B-rep) models. The major issue of processing such variational B-rep models (in the STEP format) is that free motions of a well-constrained model involve more than just rigid-body motions. The fundamental difficulty lies in having a systematic description of what pattern these free motions follow. This paper proposes a geometric perturbation method to study these free motions. This method is a generalization of the witness method, allowing it to directly deal with variational B-rep models represented by the STEP format. This generalization is essentially achieved by using a direct, geometric representation of the free motions, and then expressing the free motions in terms of composites of several basis motions. To demonstrate the effectiveness of the proposed method, a series of comparisons and case studies are presented. Highlights A new method to analyze geometric constraint systems for direct modeling. A generalization of the witness configuration method. A new method to characterize the constraint states of variational B-rep models.

Scaled boundary isogeometric analysis of large deformations in solids

AbstractThe so‐called scaled boundary isogeometric analysis combines the advantages of the isogeometric analysis and the scaled boundary finite element method. Here, the parameterization of the solid follows the idea of the scaled boundary finite element method (SB‐FEM), where the boundary of the domain is scaled in respect to a specified scaling center inside the domain. In the framework of isogeometric analysis (IGA), the NURBS functions that describe the geometry also interpolate the unknown displacement field. Such a combined approach is advantageous for three‐dimensional solids, as a radial scaling parameter describes the interior of the solid and only the geometry of the boundary is required for the analysis. The motivation is therefore to fit the idea of the boundary representation modeling technique, which is the way solids are designed in CAD –namely only by their boundary surfaces. Our contribution introduces a formulation for geometrical nonlinear 2D problems. The derived formulation addresses hyperelastic material behavior for large deformations and plane strain conditions for the two‐dimensional domain. We solve the boundary value problem by applying the weak form of equilibrium in both radial scaling and circumferential direction of the boundary. Thereafter, the nonlinear deformation behavior requires a linearization and the application of the Newton‐Raphson iterative scheme. Finally, we study the performance of this approach on a numerical problem by comparison to the standard finite element method.

Fast pose estimation for texture-less objects based on B-Rep model

Widespread industrial products, which are usually texture-less, are mainly represented with 3D boundary representation (B-Rep) model for designing and manufacturing, hence the pose estimation of texture-less object based on B-Rep model is worthy of much studying in industrial inspection. In view of such facts that surfaces are much crucial both to construction of B-Rep model and to recognition of real object, the edges of the visible surfaces in each aspect view of B-Rep model are computed and the edges in a search image containing real B-Rep objects are extracted with modified Hough algorithm. Secondly, the two edge sets are converted into the metric space for comparison, where each edge is expressed with the tetrad of edge length, angle of middle point, angle of perpendicular axis, and length of perpendicular axis. In that way, the pose of real B-Rep object in a search image is estimated by comparing the edge set of every aspect view with the edge set of the search image with the bipartite graph matching algorithm. The corresponding experiment was taken with some products in national design reservoir (NDR), and it verified the effectiveness of the texture-less pose estimation approach based on B-Rep model.

DETECTION AND EVALUATION OF TOPOLOGICAL CONSISTENCY IN CITYGML DATASETS

Abstract. The topological consistency of Boundary-Representation models, meaning here that the incidence graph is homeomorphic with the underlying topology of geographical data, is checked for several CityGML datasets, and a first classification of topological inconsistencies is performed. The analysis is carried out on a spatial database system into which the datasets have been imported. It is found that real-world datasets contain many topologically inconsistent pairs of intersecting polygons. Also data satisfying the ISO/OGC standards can still be topologically inconsistent. In the case when the intersection is a point, topological inconsistency occurs because a vertex lies on a line segment. However, the most frequent topological inconsistencies seem to arise when the intersection of two polygons is a line segment. Consequently, topological queries in present CityGML data cannot rely on the incidence graph only, but must always make costly geometric computations if correct results are to be expected.