Physics-informed neural network–integrated point cloud decimation and 3D mesh modeling for thermo-mechanical simulation of rail fasteners in digital twin systems
DOI: https://doi.org/10.3846/jcem.2026.26522Abstract
High-density point cloud data of infrastructure components, such as railway fasteners, often contain excessive noise and redundant points, making them challenging to process for simulation and analysis. This study introduces an integrated framework that transforms raw point clouds into simulation-ready 3D meshes and couples them with PhysicsInformed Neural Networks (PINNs) for thermo-mechanical analysis, enabling applications within digital twin environments. The pipeline begins with a RANSAC-based iterative segmentation and outlier removal, achieving an 82.35% reduction in point count while preserving essential geometry. Poisson surface reconstruction and targeted post-processing then produce a high-fidelity, watertight mesh normalized for consistent simulation input. Leveraging this mesh, PINNs solve the steady-state heat equation to model thermal conduction and linear elasticity equations to estimate stress and displacement fields under defined Dirichlet boundary conditions. The resulting temperature and stress distributions are visualized directly on the mesh, providing interpretable, physics-consistent insights into fastener performance. By unifying geometric simplification with data-driven, physics-aware simulation, this approach supports accurate and computationally efficient digital twin development for railway asset monitoring and maintenance.
Keywords:
digital twin, physics-informed neural networks, point cloud decimation, 3D mesh generation, rail fasteners, thermo-mechanical simulationHow to Cite
Share
License
Copyright (c) 2026 The Author(s). Published by Vilnius Gediminas Technical University.
This work is licensed under a Creative Commons Attribution 4.0 International License.
References
Agrawal, T., & Choudhary, P. (2023). Segmentation and classification on chest radiography: a systematic survey. Visual Computer, 39(3), 875–913. https://doi.org/10.1007/s00371-021-02352-7
Alexa, M., Behr, J., Cohen-Or, D., Fleishman, S., Levin, D., & Silva, C. T. (2003). Computing and rendering point set surfaces. IEEE Transactions on Visualization and Computer Graphics, 9(1), 3–15. https://doi.org/10.1109/TVCG.2003.1175093
Aljumaily, H., Laefer, D. F., & Cuadra, D. (2017). Urban point cloud mining based on density clustering and MapReduce. Journal of Computing in Civil Engineering, 31(5), Article 04017021. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000674
Arce Munoz, S. (2020). Optimized 3D reconstruction for infrastructure inspection with automated structure from motion and machine learning methods [Master’s thesis]. Brigham Young University.
Balakrishnan Selvakumaran, S., & Hall, D. M. (2022). From crowd to cloud: simplified automatic reconstruction of digital building assets for facility management. Journal of Facilities Management, 20(3), 401–436. https://doi.org/10.1108/JFM-02-2021-0017
Bassier, M., Vergauwen, M., & Poux, F. (2020). Point cloud vs. mesh features for building interior classification. Remote Sensing, 12(14), Article 2224. https://doi.org/10.3390/rs12142224
Bello, S. A., Yu, S., Wang, C., Adam, J. M., & Li, J. (2020). Review: Deep learning on 3D point clouds. Remote Sensing, 12(11), Article 1729. https://doi.org/10.3390/rs12111729
Boissonnat, J. D., & Cazals, F. (2001). Coarse-to-fine surface simplification with geometric guarantees. Computer Graphics Forum, 20(3), 490–499. https://doi.org/10.1111/1467-8659.00542
Cai, G., Zhu, Y., Wu, Y., Jiang, X., Ye, J., & Yang, D. (2023). A multimodal transformer to fuse images and metadata for skin disease classification. Visual Computer, 39(7), 2781–2793. https://doi.org/10.1007/s00371-022-02492-4
Cao, D., Wang, C., Du, M., & Xi, X. (2024a). A multiscale filtering method for airborne LiDAR data using modified 3D alpha shape. Remote Sensing, 16(8), Article 1443. https://doi.org/10.3390/rs16081443
Cao, H., Chen, D., Zhang, Y., Zhou, H., Wen, D., & Cao, C. (2024b). MFINet: a multi-scale feature interaction network for point cloud registration. The Visual Computer, 41(6), 4067–4079. https://doi.org/10.1007/s00371-024-03646-2
Chen, J., Fang, Y., Cho, Y. K., & Kim, C. (2017). Principal axes descriptor for automated construction-equipment classification from point clouds. Journal of Computing in Civil Engineering, 31(2), Article 04016058. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000628
Chen, M., Feng, A., McAlinden, R., & Soibelman, L. (2020). Photogrammetric point cloud segmentation and object information extraction for creating virtual environments and simulations. Journal of Management in Engineering, 36(2), Article 04019046. https://doi.org/10.1061/(ASCE)ME.1943-5479.0000737
Cura, R., Perret, J., & Paparoditis, N. (2017). A scalable and multi-purpose point cloud server (PCS) for easier and faster point cloud data management and processing. ISPRS Journal of Photogrammetry and Remote Sensing, 127, 39–56. https://doi.org/10.1016/j.isprsjprs.2016.06.012
Czerniawski, T., Sankaran, B., Nahangi, M., Haas, C., & Leite, F. (2018). 6D DBSCAN-based segmentation of building point clouds for planar object classification. Automation in Construction, 88, 44–58. https://doi.org/10.1016/j.autcon.2017.12.029
de la Plata, A. R. M., Franco, P. A. C., Franco, J. C., & Bravo, V. G. (2021). Protocol development for point clouds, triangulated meshes and parametric model acquisition and integration in an hbim workflow for change control and management in a unesco’s world heritage site. Sensors, 21(4), Article 1083. https://doi.org/10.3390/s21041083
Dekker, B., Ton, B., Meijer, J., Bouali, N., Linssen, J., & Ahmed, F. (2023). Point Cloud analysis of railway infrastructure: A systematic literature review. IEEE Access, 11, 134355–134373. https://doi.org/10.1109/ACCESS.2023.3337049
Dey, T. K., & Goswami, S. (2003). Tight cocone: A water-tight surface reconstructor. Journal of Computing and Information Science in Engineering, 3(4), 302–307. https://doi.org/10.1115/1.1633278
Dey, T. K., Giesen, J., & Goswami, S. (2003). Shape segmentation and matching with flow discretization. In F. Dehne, J. R. Sack, & M. Smid (Eds.), Lecture notes in computer science: Vol. 2748. Algorithms and data structures. WADS 2003 (pp. 25–36). Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-540-45078-8_3
Fathi, H., & Brilakis, I. (2011). Automated sparse 3D point cloud generation of infrastructure using its distinctive visual features. Advanced Engineering Informatics, 25(4), 760–770. https://doi.org/10.1016/j.aei.2011.06.001
Franco-Duran, D. M., & Mejia, A. G. (2016). Assessing the cost forecasting performance in construction projects through data envelopment analysis. In Proceedings of Construction Research Congress 2016 (pp. 2039–2049). ASCE. https://doi.org/10.1061/9780784479827.203
Gao, M., Cao, T.-T., Tan, T., & Huang, Z. (2011). gHull: a three-dimensional convex hull algorithm for graphics hardware. In I3D ‘11: Symposium on Interactive 3D Graphics and Games (pp. 204), California, San Francisco. https://doi.org/10.1145/1944745.1944784
Hajihosseinlou, M., Maghsoudi, A., & Ghezelbash, R. (2024). Intelligent mapping of geochemical anomalies: Adaptation of DBSCAN and mean-shift clustering approaches. Journal of Geochemical Exploration, 258, Article 107393. https://doi.org/10.1016/j.gexplo.2024.107393
Hidaka, J., Elfeki, H., Duelund-Jakobsen, J., Laurberg, S., & Lundby, L. (2019). Functional outcome after laparoscopic posterior sutured rectopexy versus ventral mesh rectopexy for rectal prolapse: six-year follow-up of a double-blind, randomized single-center study. eClinicalMedicine, 16(12), 18–22. https://doi.org/10.1016/j.eclinm.2019.08.014
Hinks, T., Carr, H., Truong-Hong, L., & Laefer, D. F. (2013). Point cloud data conversion into solid models via point-based voxelization. Journal of Surveying Engineering, 139(2), 72–83. https://doi.org/10.1061/(ASCE)SU.1943-5428.0000097
Hou, G., Wang, J., & Fan, Y. (2024). Wind power forecasting method of large-scale wind turbine clusters based on DBSCAN clustering and an enhanced hunter-prey optimization algorithm. Energy Conversion and Management, 307, Article 118341. https://doi.org/10.1016/j.enconman.2024.118341
Hu, Q., Yang, B., Xie, L., Rosa, S., Guo, Y., Wang, Z., Trigoni, N., & Markham, A. (2022). Learning semantic segmentation of large-scale point clouds with random sampling. IEEE Transactions on Pattern Analysis and Machine Intelligence, 44(11), 8338–8354. https://doi.org/10.1109/TPAMI.2021.3083288
Huang, Z., Zheng, H., Li, C., & Che, C. (2024). Application of machine learning-based k-means clustering for financial fraud detection. Academic Journal of Science and Technology, 10(1), 33–39. https://doi.org/10.54097/74414c90
Jing, H., Meng, X., Slatcher, N., & Hunter, G. (2021). Efficient point cloud corrections for mobile monitoring applications using road/rail-side infrastructure. Survey Review, 53(378), 235–251. https://doi.org/10.1080/00396265.2020.1719753
Justo, A., Soilán, M., Sánchez-Rodríguez, A., & Riveiro, B. (2021). Scan-to-BIM for the infrastructure domain: Generation of IFC-complaint models of road infrastructure assets and semantics using 3D point cloud data. Automation in Construction, 127, Article 103703. https://doi.org/10.1016/j.autcon.2021.103703
Kazhdan, M., Bolitho, M., & Hoppe, H. (2006). Poisson surface reconstruction. In Eurographics Symposium on Geometry Processing, Cagliari, Sardinia, Italy. https://hhoppe.com/poissonrecon.pdf
Li, Z., Cheng, C., Kwan, M. P., Tong, X., & Tian, S. (2019). Identifying asphalt pavement distress using UAV LiDAR point cloud data and random forest classification. ISPRS International Journal of Geo-Information, 8(1), Article 39. https://doi.org/10.3390/ijgi8010039
Li, X., Zhang, Q., Kang, D., Cheng, W., Gao, Y., Zhang, J., Liang, Z., Liao, J., Cao, Y.-P., & Shan, Y. (2024). Advances in 3D generation: A survey. ArXiv. https://doi.org/10.48550/arXiv.2401.17807
Liu, J., Xiao, Z., Lu, S., Che, D., Dong, M., & Bai, C. (2023). Infrastructure-level support for GPU-enabled deep learning in DATAVIEW. Future Generation Computer Systems, 141, 723–737. https://doi.org/10.1016/j.future.2022.12.014
Mansour, M., Martens, J., & Blankenbach, J. (2024). Hierarchical SVM for semantic segmentation of 3D point clouds for infrastructure scenes. Infrastructures, 9(5), Article 83. https://doi.org/10.3390/infrastructures9050083
Mirzaei, K., Arashpour, M., Asadi, E., Masoumi, H., Bai, Y., & Behnood, A. (2022). 3D point cloud data processing with machine learning for construction and infrastructure applications: A comprehensive review. Advanced Engineering Informatics, 51, Article 101501. https://doi.org/10.1016/j.aei.2021.101501
Moenning, C., & Dodgson, N. a. (2004). Intrinsic point cloud simplification. In Proceedings of International Conference Graphicon 2004 (pp. 6–10), Moscow, Russia.
Pauly, M., Gross, M., & Kobbelt, L. P. (2002). Efficient simplification of point-sampled surfaces. In Proceedings of the IEEE Visualization Conference (VIS 2002) (pp. 163–170), Boston, MA, USA. IEEE. https://doi.org/10.1109/VISUAL.2002.1183771
Qiu, S., Zaheer, Q., Ehsan, H., Shah, S. M. A. H., Ai, C., Wang, J., & Zheng, A. A. (2024). Multimodal fusion network for crack segmentation with modified U-Net and transfer learning – based MobileNetV2. Journal of Infrastructure Systems, 30(4), Article 04024029. https://doi.org/10.1061/JITSE4.ISENG-2499
Raj, C., Khular, L., & Raj, G. (2020). Clustering based incident handling for anomaly detection in cloud infrastructures. In Proceedings of the 2020 10th International Conference on Cloud Computing, Data Science & Engineering (Confluence) (pp. 611–616), Noida, India. IEEE. https://doi.org/10.1109/Confluence47617.2020.9058314
Rashidi, A., Brilakis, I., & Vela, P. A. (2013). Built infrastructure point cloud data cleaning: an overview of gap filling algorithms. In Proceedings of the 13th International Conference on Construction Applications of Virtual Reality (pp. 585–593), London, UK.
Rausch, C., & Haas, C. (2021). Automated shape and pose updating of building information model elements from 3D point clouds. Automation in Construction, 124, Article 103561. https://doi.org/10.1016/j.autcon.2021.103561
Richter, R., & Döllner, J. (2014). Concepts and techniques for integration, analysis and visualization of massive 3D point clouds. Computers, Environment and Urban Systems, 45, 114–124. https://doi.org/10.1016/j.compenvurbsys.2013.07.004
Sakr, S., Liu, A., Batista, D. M., & Alomari, M. (2011). A survey of large scale data management approaches in cloud environments. IEEE Communications Surveys and Tutorials, 13(3), 311–336. https://doi.org/10.1109/SURV.2011.032211.00087
Saovana, N., Yabuki, N., & Fukuda, T. (2021). Automated point cloud classification using an image-based instance segmentation for structure from motion. Automation in Construction, 12, Article 103804. https://doi.org/10.1016/j.autcon.2021.103804
Soilán, M., Justo, A., Sánchez-Rodríguez, A., Lamas, D., & Riveiro, B. (2021). 3D point cloud data processing and infrastructure information models: Methods and findings from safeway project. International Archives of the Photogrammetry, Remote Sensing and Spatial Information Sciences, 43(B2-2021), 239–246. https://doi.org/10.5194/isprs-archives-XLIII-B2-2021-239-2021
Taubin, G. (1995). Curve and surface smoothing without shrinkage. In Proceedings of IEEE International Conference on Computer Vision (pp. 852–857), Cambridge, MA, USA. IEEE. https://doi.org/10.1109/ICCV.1995.466848
Vassilev, H., Laska, M., & Blankenbach, J. (2024). Uncertainty-aware point cloud segmentation for infrastructure projects using Bayesian deep learning. Automation in Construction, 164, Article 105419. https://doi.org/10.1016/j.autcon.2024.105419
Vick, S., & Brilakis, I. (2018). Road design layer detection in point cloud data for construction progress monitoring. Journal of Computing in Civil Engineering, 32(5), Article 0401802. https://doi.org/10.1061/(ASCE)CP.1943-5487.0000772
Walsh, S. B., Borello, D. J., Guldur, B., & Hajjar, J. F. (2013). Data processing of point clouds for object detection for structural engineering applications. Computer-Aided Civil and Infrastructure Engineering, 28(7), 495–508. https://doi.org/10.1111/mice.12016
Wang, J., Ahmed, S. M., Shah, H., Atta, Z., & Wang, W. (2025a). An efficient PointNet-based multifaceted autoencoder (PMAE) for denoising rail track fastener point clouds. Journal of Civil Structural Health Monitoring, 15, 3995–4015. https://doi.org/10.1007/s13349-025-01026-5
Wang, J., Ahmed, S. M., Shah, H., Ehsan, H., Faizan, S., Shah, H., Ai, C., Kuang, J., Wang, W., & Qiu, S. (2025b). Self supervised contrastive anomaly detection in railway fasteners using point clouds and deep metric learning for imbalance dataset. Journal of Civil Structural Health Monitoring, 15, 2861–2886. https://doi.org/10.1007/s13349-025-00960-8
Wang, W., Ehsan, H., Qiu, S., Rahman, T. U., Wang, J., & Zaheer, Q. (2026). Evolution and emerging frontiers in point cloud technology. Electronics, 15(2), Article 341. https://doi.org/10.3390/electronics15020341
Weerasinghe, U. (2023). LIDAR-based 3D object detection for autonomous driving a comprehensive exploration of methods, implementation steps, tools, and challenges in integrating deep learning techniques. International Journal of Sustainable Infrastructure for Cities and Societies, 8(2), 52–64.
Xu, J., Xu, K., Wang, Y., Shen, Q., & Li, R. (2024). A K-means algorithm for financial market risk forecasting. ArXiv. https://doi.org/10.48550/arXiv.2405.13076
Yuan, X., Chen, H., & Liu, B. (2021). Point cloud clustering and outlier detection based on spatial neighbor connected region labeling. Measurement and Control, 54(5–6), 835–844. https://doi.org/10.1177/0020294020919869
Zaheer, Q., Ahmed, S. M., Shah, H., Wang, W., Ehsan, H., Ai, C., Wang, J., & Qiu, S. (2026a). Multimodal graph neural network framework for railway fastener tightness assessment from high-resolution point clouds. Engineering Applications of Artificial Intelligence, 167(P2), Article 113829. https://doi.org/10.1016/j.engappai.2026.113829
Zaheer, Q., Ehsan, H., Wang, W., Malik, M., & Wang, J. (2026b). Real-time synthetic data generation and segmentation of railway fasteners using an attention guided dual-phase diffusion-aided model. Measurement, 261, Article 119874. https://doi.org/10.1016/j.measurement.2025.119874
Zaheer, Q., Ehsan, H., Wang, W., Shah, S. F. H., Ai, C., Wang, J., & Qiu, S. (2026c). DA-NGF: A domain-adaptive neurogenerative framework for few-shot railway fastener defect synthesis and transferable representation learning. Applied Intelligence, 56, Article 32. https://doi.org/10.1007/s10489-025-07054-4
Zhang, G., Vela, P. A., Karasev, P., & Brilakis, I. (2015). A sparsity-inducing optimization-based algorithm for planar patches extraction from noisy point-cloud data. Computer-Aided Civil and Infrastructure Engineering, 30(2), 85–102. https://doi.org/10.1111/mice.12063
Zhang, J., Zhao, X., Chen, Z., & Lu, Z. (2019). A review of deep learning-based semantic segmentation for point cloud. IEEE Access, 7, 179118–179133. https://doi.org/10.1109/ACCESS.2019.2958671
Zhuang, H., Zhang, J., & Liao, F. (2023). A systematic review on application of deep learning in digestive system image processing. Visual Computer, 39(6), 2207–2222. https://doi.org/10.1007/s00371-021-02322-z
View article in other formats
CrossMark check
Published
Issue
Section
Copyright
Copyright (c) 2026 The Author(s). Published by Vilnius Gediminas Technical University.
License
This work is licensed under a Creative Commons Attribution 4.0 International License.