Computer-Aided Design最新文献

{"title":"MidSurfer: Efficient Mid-Surface Abstraction from Variable Thin-Walled Models","authors":"Li Ye ,&nbsp;Xinhang Zhou ,&nbsp;Peng Fan ,&nbsp;Ruofeng Tong ,&nbsp;Hailong Li ,&nbsp;Peng Du ,&nbsp;Min Tang","doi":"10.1016/j.cad.2025.103965","DOIUrl":"10.1016/j.cad.2025.103965","url":null,"abstract":"<div><div>This paper addresses the challenge of efficiently abstracting mid-surfaces from complex variable thin-walled models, a critical task in computer-aided design (CAD) and finite element analysis (FEA) for simplifying thin-walled structures. Traditional methods often require manual specification of pairing faces, which can be time-consuming and error-prone. Alternatively, automatic face pairing methods fail to meet the actual needs of variable thin-walled models, resulting in the accumulation of topological errors. Additionally, existing algorithms struggle to extract mid-surfaces from models with varying wall thickness or produce mid-surfaces with poor accuracy, leading to geometric errors. Furthermore, the computational efficiency of these methods is often inadequate for large-scale models. To overcome these challenges, we propose an automated face-pairing mechanism that eliminates the need for manual intervention, enhancing the algorithm’s robustness and enabling it to handle cases that the commercial CAD modeling engine, Parasolid, cannot process. Our approach accurately processes variable thin-walled models, with results closely aligning with the ground truth, as demonstrated by the provided error distribution tables. Moreover, our algorithm achieves a <span><math><mrow><mn>4</mn><mo>−</mo><mn>12</mn></mrow></math></span> times improvement in efficiency than previous methods over the geometry extraction stage and supports multi-threaded acceleration, significantly reducing computation time. Experimental results demonstrate that our algorithm surpasses existing methods in both accuracy and efficiency, offering a promising solution for mid-surface extraction in complex, variable thin-walled models.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103965"},"PeriodicalIF":3.1,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159994","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Interactive design of developable surfaces by patch-based learning","authors":"Chaoyun Wang ,&nbsp;Jianlei Wang ,&nbsp;Chengcheng Tang ,&nbsp;Nanning Zheng ,&nbsp;Caigui Jiang","doi":"10.1016/j.cad.2025.103970","DOIUrl":"10.1016/j.cad.2025.103970","url":null,"abstract":"<div><div>This paper introduces an interactive design method for developable surfaces, centered on a data-driven approach to optimize surface patches for developability. Surface patches are the fundamental components of an entire surface, typically represented by triangular meshes. We propose a novel learning-based method that effectively transforms patches with arbitrary boundaries into their closest developable surfaces. Based on this method, our tools enable real-time, drag-and-drop design of developable surfaces and support piecewise developable approximation through interactive inputs. Experimental results demonstrate that this method provides a fast computational foundation for the interactive design of developable surfaces, enhancing design flexibility while exhibiting excellent robustness and generalization. The piecewise developable approximation of the model, guided by human–computer collaborative segmentation, achieved higher overall approximation accuracy, fewer patches, and lifelike papercraft outcomes. This offers greater flexibility to meet the application requirements of complex real-world scenarios and provides a new paradigm for integrating deep learning with interactive geometry design.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103970"},"PeriodicalIF":3.1,"publicationDate":"2025-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159992","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"On design, analysis, and hybrid manufacturing of microstructured blade-like geometries","authors":"Pablo Antolin ,&nbsp;Michael Bartoň ,&nbsp;Georges-Pierre Bonneau ,&nbsp;Annalisa Buffa ,&nbsp;Amaia Calleja-Ochoa ,&nbsp;Gershon Elber ,&nbsp;Stefanie Elgeti ,&nbsp;Gaizka Gómez Escudero ,&nbsp;Alicia Gonzalez ,&nbsp;Haizea González Barrio ,&nbsp;Stefanie Hahmann ,&nbsp;Thibaut Hirschler ,&nbsp;Q Youn Hong ,&nbsp;Konstantin Key ,&nbsp;Myung-Soo Kim ,&nbsp;Michael Kofler ,&nbsp;Norberto Lopez de Lacalle ,&nbsp;Silvia de la Maza ,&nbsp;Kanika Rajain ,&nbsp;Jacques Zwar","doi":"10.1016/j.cad.2025.103967","DOIUrl":"10.1016/j.cad.2025.103967","url":null,"abstract":"<div><div>With the evolution of new manufacturing technologies such as multi-material 3D printing, one can think of new type of objects that consist of considerably less, yet heterogeneous, material, consequently being porous, lighter and cheaper, while having the very same functionality as the original object when manufactured from one single solid material. We aim at questioning five decades of traditional paradigms in geometric CAD and focus at new generation of CAD objects that are not solid, but contain heterogeneous free-form internal microstructures. We propose a unified manufacturing pipeline that involves all stages, namely design, optimization, manufacturing, and inspection of microstructured free-form geometries. We demonstrate our pipeline on an industrial test case of a blisk blade that sustains the desired pressure limits, yet requires significantly less material when compared to the solid counterpart.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103967"},"PeriodicalIF":3.1,"publicationDate":"2025-09-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120121","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Size-controlled quadrilateral meshing using integrable odeco fields","authors":"Mattéo Couplet ,&nbsp;Alexandre Chemin ,&nbsp;Jean-François Remacle","doi":"10.1016/j.cad.2025.103974","DOIUrl":"10.1016/j.cad.2025.103974","url":null,"abstract":"<div><div>This paper proposes a novel approach for computing planar quadrilateral meshes complying with sizing prescriptions on boundary and feature curves. The method relies on computing <em>integrable</em> orthogonal frame fields, whose symmetries are implicitly represented using orthogonally decomposable (<em>odeco</em>) tensors. To formulate an integrability criterion, we express the frame field’s Lie bracket solely in terms of the tensor representation; this is made possible by studying the sensitivity of the frame with respect to perturbations in the tensor. We construct an energy formulation that computes smooth and integrable frame fields in both isotropic and anisotropic settings. The solver creates and places the singularities required to fit the sizing constraints with the correct topology. The computed frame field is integrated to a seamless parametrization that is aligned with the frame field, and we propose a mesh extraction method that relies on a greedy quantization of the parametrization.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103974"},"PeriodicalIF":3.1,"publicationDate":"2025-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145159993","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Unfolding 3D shape convolutional networks for aerodynamic drag prediction in vehicle design","authors":"Yangchen Liu ,&nbsp;Yubin Tang ,&nbsp;Xiaomeng Tong ,&nbsp;Qinglei Cao ,&nbsp;Zichuan Fan","doi":"10.1016/j.cad.2025.103971","DOIUrl":"10.1016/j.cad.2025.103971","url":null,"abstract":"<div><div>The aerodynamic drag coefficient is crucial in vehicle design, especially during the shape design phase, where the exterior geometry directly affects the drag coefficient. Reducing the drag coefficient enhances fuel efficiency, lowers emissions, and improves vehicle dynamics and stability. Nevertheless, existing methods like Computational Fluid Dynamics (CFD) are time-consuming and computationally intensive, making them unsuitable for real-time feedback. To address this, we propose an innovative deep learning model — the Unfolded Convolutional Neural Network (UFConv). The ABTrans and Distance Mapping method within UFConv reduces complexity by unfolding the 3D structure into a lower-dimensional representation. This accelerates training while maintaining high accuracy by extracting key features and avoiding redundant information. The R<span><math><msup><mrow></mrow><mrow><mn>2</mn></mrow></msup></math></span> values for the UFConv model are 0.941 and 0.829 on the DrivAerNet and DrivAerNet++ datasets, respectively, with a peak of 0.952 for a specific vehicle type. The training time is also significantly lower compared to other methods. The results indicate that UFConv outperforms existing methods, providing superior accuracy and faster training speed in predicting automotive drag.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103971"},"PeriodicalIF":3.1,"publicationDate":"2025-09-16","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145120119","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Mesh generation of curvilinear polygons for the high-order virtual element method (VEM)","authors":"Kaloyan S. Kirilov ,&nbsp;Jingtian Zhou ,&nbsp;Joaquim Peiró ,&nbsp;Mashy Green ,&nbsp;David Moxey ,&nbsp;Lourenço Beirão da Veiga ,&nbsp;Alessandro Russo ,&nbsp;Franco Dassi","doi":"10.1016/j.cad.2025.103966","DOIUrl":"10.1016/j.cad.2025.103966","url":null,"abstract":"<div><div>We present a proof-of-concept methodology for generating curvilinear polygonal meshes suitable for high-order discretisations by the Virtual Element Method (VEM). A VEM discretisation requires the definition of a set of boundary and internal points used to define basis functions and compute integrals of polynomials. The procedure to locate these points on the boundary borrows ideas from previous work on <em>a posteriori</em> high-order mesh generation in which the geometrical inquiries to a B-rep model of the computational domain are performed via an interface to CAD libraries.</div><div>Here we describe the steps of the procedure that transforms a straight-sided polygonal mesh, generated using third-party software, into a curvilinear boundary-conforming mesh. We discuss criteria for ensuring and verifying the validity of the mesh. Using an elliptic partial differential equation with Dirichlet boundary conditions as a model problem, we show that VEM discretisations on such meshes achieve the expected rates of convergence as the mesh resolution is increased. This is followed by an illustrative application of the method to the generation of a curvilinear polygonal mesh for an aerofoil geometry.</div><div>We discuss polygonal curvilinear mesh quality and its enhancement, and use the motion of a cell vertex to appraise three elemental quality metrics, namely convexity, regularity and isotropy, and highlight some of the difficulties associated in their use for mesh quality optimisation. A derivative-free optimisation method is utilised to enhance curvilinear polygonal meshes by maximising a suitable measure of mesh quality. We propose such measure as a combination of the three quality metrics and apply it to optimise a distorted initial mesh for a ring geometry. We show that a suitable version of the convexity metric is effective in untangling invalid meshes. The VEM solution of a model elliptic equation is obtained for a ring geometry where a distorted and an optimised mesh show low errors, indicating that the VEM is robust and relatively insensitive to mesh distortion, and a reduction of the error in the optimised mesh.</div><div>Finally, we use a more complex geometry, a computational domain for an aerofoil, as a benchmark to further illustrate the ability of the convexity metric to untangle meshes, and also to assess the suitability of two quality measures as optimisation targets to improve the overall quality of curvilinear polygonal meshes.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103966"},"PeriodicalIF":3.1,"publicationDate":"2025-09-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145099147","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Quadrilateral mesh generation based on foliation and meromorphic quadratic differential","authors":"Xiaopeng Zheng,&nbsp;Hao Wang,&nbsp;Na Lei,&nbsp;Zhongxuan Luo","doi":"10.1016/j.cad.2025.103953","DOIUrl":"10.1016/j.cad.2025.103953","url":null,"abstract":"<div><div>Quadrilateral meshes derived from foliations and quadratic differentials possess a high structural regularity. However, for complex models, meshes directly generated by foliation and its induced holomorphic quadratic differentials face notable challenges regarding area distortion, corner preservation, and uniform cell size distribution. To overcome these limitations, we introduce a set of enhanced techniques grounded in surface foliation and meromorphic quadratic differentials. Specifically, we introduce pole-constrained foliations to compute meromorphic quadratic differentials, significantly reducing area distortion. Additionally, a modified double cover strategy is further introduced to preserve corner features by altering the model’s topology. Finally, adaptive metric graph optimization is utilized to ensure a uniform distribution of mesh elements. Experiments validate the effectiveness of the proposed approach.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"190 ","pages":"Article 103953"},"PeriodicalIF":3.1,"publicationDate":"2025-09-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145099143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Principal stress field-guided optimization for rib structure generation","authors":"Longdu Liu ,&nbsp;Xiangjun Wu ,&nbsp;Jiqiang Huang ,&nbsp;Lingxin Cao ,&nbsp;Xiaokang Liu ,&nbsp;Changhe Tu ,&nbsp;Lin Lu","doi":"10.1016/j.cad.2025.103955","DOIUrl":"10.1016/j.cad.2025.103955","url":null,"abstract":"<div><div>Shell structures are widely used in architecture and engineering for their efficient load-bearing behavior. However, large-span, thin-shell designs often suffer from insufficient stiffness and strength. Rib-reinforced shells can enhance both stiffness and spatial efficiency, but they commonly rely on inefficient principal-stress-field (PSF)–guided quadrilateral partitioning. We present a direct computational framework that optimizes rib layouts from the PSF without intermediate partitioning. Our approach uses centroidal Voronoi tessellation to generate ribs directly on the optimized principal stress field, enabling adaptive rib refinement on arbitrary freeform surfaces. We also optimize rib cross-sectional profiles to minimize material usage while preserving structural performance, improving both mechanical efficiency and sustainability. Numerical simulations and physical experiments show that, under equivalent load and volume constraints, structures optimized with our method achieve a 78% reduction in deformation compared to conventional approaches, validating the effectiveness of the framework.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"189 ","pages":"Article 103955"},"PeriodicalIF":3.1,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145048984","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Optimization of a Triangular Delaunay Mesh Generator using Reinforcement Learning","authors":"Will Thacher ,&nbsp;Yulong Pan ,&nbsp;Per-Olof Persson","doi":"10.1016/j.cad.2025.103964","DOIUrl":"10.1016/j.cad.2025.103964","url":null,"abstract":"<div><div>In this work we introduce a triangular Delaunay mesh generator that can be trained using reinforcement learning to maximize a given mesh quality metric. Our mesh generator consists of a graph neural network that distributes and modifies vertices, and a standard Delaunay algorithm to triangulate the vertices. We explore various design choices and evaluate our mesh generator on diverse tasks including mesh generation, mesh improvement, and producing variable resolution meshes. The learned mesh generator outputs meshes that are comparable in quality to those produced by Triangle and DistMesh, two popular Delaunay-based mesh generators.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"189 ","pages":"Article 103964"},"PeriodicalIF":3.1,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988420","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"MCCMS: Achieve fine-grained phase distribution design in cement microstructure using diffusion models","authors":"Jiahao Li ,&nbsp;Xin Zhao ,&nbsp;Lin Wang ,&nbsp;Shuangrong Liu ,&nbsp;Haozhong Gao ,&nbsp;Zeming Cheng ,&nbsp;Chaoran Pang ,&nbsp;Bo Yang","doi":"10.1016/j.cad.2025.103946","DOIUrl":"10.1016/j.cad.2025.103946","url":null,"abstract":"<div><div>The design of phase distribution in cement microstructure has significant research and educational value. It aids in correlation studies, expands knowledge boundaries, and guides production processes. However, current methods for controllable microstructure synthesis cannot customize the location, size, and shape of phases in the synthesized microstructure, limiting research on cement-based materials. To address this limitation, this paper proposes a diffusion-based synthesis framework, MCCMS, which introduces multi-scale constraints at different diffusion steps to achieve fine-grained customization of phases. The MCCMS framework divides the cement microstructure into super-voxels of varying sizes at each diffusion step, imposing constraint rules to regulate their variation. Users can modify these super-voxels at different steps to customize the phases under varying constraint strengths. Experimental results demonstrate that the MCCMS framework can precise control over the position, size, and shape of phases within the microstructure, showcasing the high fidelity of the synthetic results.</div></div>","PeriodicalId":50632,"journal":{"name":"Computer-Aided Design","volume":"189 ","pages":"Article 103946"},"PeriodicalIF":3.1,"publicationDate":"2025-09-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144988421","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}