A Skeleton-Tree-Based Structural Retrieval Method for Spherical Hybrid Sliding Bearings

DOI: https://doi.org/10.62517/jes.202602134

Author(s)

Fangting Liu1, Yawen Fan2,*, Jingfeng Shen1,*, Zichuan Wang1, Shikun Zhang1

Affiliation(s)

1School of Mechanical Engineering, University of Shanghai for Science and Technology, Shanghai , China 2Sino-British International College, University of Shanghai for Science and Technology, Shanghai, China *Corresponding Author

Abstract

To address the complex internal fluid-domain structure of Spherical Hybrid Sliding Bearings (SHSBs) and the inability of traditional geometric retrieval methods to capture topological features, this paper proposes an intelligent retrieval method based on skeleton trees. The bearing fluid-domain model is extracted by Boolean operations and uniformly voxelized. A topologically equivalent centerline skeleton is then extracted using the Euclidean distance transform and gradient vector flow (GVF) algorithm. Key feature nodes of the skeleton are identified using a 3 x 3 x 3 neighborhood convolution operator, and a four-dimensional global feature vector is constructed from the volume fraction, normalized skeleton length, number of endpoints, and number of branch points. Cosine similarity is finally used to achieve quantitative retrieval of bearing structures. Validation results show that the proposed method achieves a similarity of up to 99.9% for closely similar structures, effectively identifying geometric measure consistency and local topological differences.

Keywords

Spherical Hybrid Sliding Bearing; Structural Retrieval; Skeleton Tree; Fluid-Domain Topology; Gradient Vector Flow; Cosine Similarity

References

[1] Li Y M. Ultra-high precision liquid hydrostatic spherical bearing system. Acta Metrologica Sinica, 1986(03):43-47. [2] Ji D S, Shen J F, Chen Y F, et al. Dynamic characteristic analysis of spherical hybrid sliding bearings. Journal of Mechanical Strength, 2022, 44(2):1-8. [3] Lu B, Fan X M. Research on 3D point cloud skeleton extraction based on improved adaptive k-means clustering. Acta Automatica Sinica, 2022, 48(8):1994-2006. [4] Fan R J, Liu J, Yu J M, et al. Small sample 3D point cloud target recognition methodbased on shape and skeleton feature matching. Systems Engineering and Electronics, 2025, 48(1):76-86. [5] Menten M J, Paetzold J C,Zimmer V A, et al. A skeletonization algorithm for gradient-based optimization//Proceedings of the IEEE/CVF International Conference on Computer Vision. 2023:21394-21403. [6] Heryan K , Caliman T D . A Novel Skeletonization Algorithm for Topologically Complex Structures: Comparative Analysis and Application to Renal Arterial Trees. IEEE Access, 2025, 13(13):134989-135006. [7] Rao Fu, Yunchi Zhang, Jie Yang, et al. ROSA-Net: Rotation-Robust Structure-Aware Network forFine-Grained 3D Shape Retrieval. Lecture Notes in Computer Science, 2024:295-319. [8] Guan Y, Kwan D, Liang R, et al. INRet: A General Framework for Accurate Retrieval of INRs for Shapes//2025 International Conference on 3D Vision (3DV). IEEE, 2025:905-914. [9] Anderson Soares da Costa Azevêdo, Li H, Ishida N ,et al. Body-fitted topology optimization via integer linear programming using surface capturing techniques. International journal for Numerical Methods in Engineering, 2024, 125(13):25. [10] Wang G, Laga H, Srivastava A. Elastic shape analysis of tree-like 3d objects using extended srvf representation. IEEE transactions on pattern analysis and machine intelligence, 2023, 46(4):2475-2488. [11] Zhu W B, Geng G Q, Liu Y Y, et al. 3D Model Retrieval Method of Mechanical Parts Based on Skeleton Tree. Journal of Mechanical Engineering, 2016, 52(13):204-212. [12] Zhang X, Liu Y Y, Yang D. Three-dimensional Model Based on Mechanical Parts Skeleton Similarity Measure Algorithm. Nonferrous Metals Materials and Engineering, 2017, 38(4):234-238. [13] Wang L, Liu J, Zheng L, et al. Meet JEANIE: A Similarity Measure for 3D Skeleton Sequences via Temporal-Viewpoint Alignment. International Journal of Computer Vision, 2024, 132(9):4091-4122. [14] Ye F L. An Improved Image Skeleton Extraction Algorithm. Journal of Xichang University (Natural Science Edition), 2018, 32(3):91-93+123.