Bridge & Tunnel logo Bridge & Tunnel Engineering Hanoi University of Civil Engineering
Research area 04

Artificial intelligence and digital twins

Machine learning, neural networks, deep learning and symbolic regression for damage detection, prediction of material properties and digital twins of structures.

26 publications26 papers2019⁠–⁠202610 with publications4 listed on staff profiles2 projects

Overview

Within the group, artificial intelligence serves two main problems: detecting structural damage from measured data, and predicting material properties from mix composition. A distinctive feature is the preference for explicit (“white-box”) formulas from genetic programming and symbolic regression that engineers can use directly in design [19, 20, 24].

Main research themes

Damage detection with neural networks and deep learning

Transmissibility functions (ratios of the response spectra at two measurement points) combined with neural networks and machine learning were applied to damage detection on the Ca-Non Bridge and the Nam O truss bridge [2, 3]. Gapped-smoothed modal curvatures were combined with convolutional neural networks and deep learning and applied to the Bo Nghi girder bridge; for slabs, two-dimensional curvature mode shapes were combined with Faster R-CNN [4, 6, 14]. Feed-forward neural networks were combined with global optimisation algorithms to detect and quantify damage [7, 8].

Digital twins

A digital twin framework combines a finite element model, model updating, signal processing and a MobileNetV2 network to detect low-severity damage in a slab from deflection data [21].

Prediction of material properties

The compressive strength and effective porosity of pervious concrete were predicted with XGBoost, random forests and support vector machines on a dataset of 164 samples [10]. Genetic programming gave explicit formulas for the compressive strength of steel-fibre-reinforced concrete and the permeability of pervious concrete; for prestressed hollow-core slabs, shear strength was predicted with data-driven models [17, 19, 20].

Projects

A university project on damage detection from vibration response and artificial neural networks, and a ministry-level project (2023⁠–⁠2024) on predicting the behaviour of pervious cement concrete with artificial intelligence and data-driven methods, belong to this area.

Research projects (2)

Project titles as on the staff profiles (Vietnamese).

  • Principal investigator (2023⁠–⁠2024). Nghiên cứu xây dựng mô hình dự báo ứng xử cơ học của bê tông xi măng rỗng dựa trên tiếp cận lý thuyết kết hợp kỹ thuật trí tuệ nhân tạo và định hướng dữ liệu. Code: B2023-XDA-03. Ministry-level project, Ministry of Education and Training (đã nghiệm thu)
  • Nghiên cứu phương pháp phát hiện và đánh giá hư hỏng của kết cấu dựa vào ứng xử dao động và mạng Nơ ron nhân tạo. Đại học Xây dựng

Directions of development

The directions below are those shown by the papers of 2023⁠–⁠2026, the projects and the research interests stated on staff profiles; this page states no plans or funding.

  1. Deep learning for non-destructive testing

    A 2026 paper uses ground-penetrating radar (GPR) with YOLOv11 to detect reinforcement corrosion in RC slabs in a fully non-destructive and automated way [22].

  2. Neural networks with genetic algorithms for construction

    A 2026 ANN–GA framework, trained on 675 simulated cases, predicts and optimises cooling-pipe parameters to control the temperature field and prevent cracking in mass concrete pile caps [26].

  3. Explicit formulas instead of black-box models

    Constrained symbolic regression for the modulus of highly filled suspensions and an explicit formula for aggregate packing density (2026) continue the move to turn machine learning into formulas [24, 25]. Staff profiles list “building digital twin models of structures” and “AI and IoT in structural monitoring” as research interests.

People in this area

With publications in this area (10)

Staff who list this area on their profile (4)

Current staff who list this area on their profile; the same group appears when the people page is filtered by area and on the area cards.

Publications in this area (26)

Every journal paper, conference paper, book chapter and book classified into this area, newest first. Members of the group in bold, * corresponding author; titles in their original language.

2026

  1. [26]Dinh-Manh Le, Van-Minh Le*, Thai-Son Vu, Huong-Duong Nguyen (2026). Artificial intelligence-based prediction of cooling pipe system parameters for thermal field control and crack prevention in mass concrete structures. Structural Concrete, 1-30. https://doi.org/10.1002/suco.70818
  2. [25]Vu Thai Son, Sengaloun Keoalounxay, Nguyen Ngoc Tuyen, Tran Viet Hung (2026). Predicting the relative shear storage modulus of highly filled bidisperse particulate suspensions by machine learning and constrained symbolic regression. Journal of Science and Technology in Civil Engineering (JSTCE) - HUCE, 20(3), 127-146. https://doi.org/10.31814/stce.huce2026-20(3)-07
  3. [24]Thai-Son Vu, Sengaloun Keoalounxay, Quoc-Bao Nguyen* (2026). An Explicit Formula for the Dry Packing Density of Aggregate Gradations Derived from the de Larrard Compressible Packing Model by Symbolic Regression. Eng, 7(9), 493. https://doi.org/10.3390/eng7090493
  4. [23]Quy, N. D.*, & Le, B. D. (2026). Real-time vibration monitoring and anomaly detection using autoencoder and wireless IoT sensor node. In Lecture Notes in Networks and Systems (pp. 260-267). Springer, Cham. https://doi.org/10.1007/978-3-032-18316-3_29
  5. [22]Bui, H. T.*, Tan, K. H.* (2026). Intelligent corrosion detection in reinforced concrete slabs using GPR and YOLOv11: A fully non-destructive and automated deep learning framework. Construction and Building Materials, 506, 145014. https://doi.org/10.1016/j.conbuildmat.2025.145014

2025

  1. [21]Nguyen, H. D., Nguyen, H., & Gao, X. (2025). A digital twin framework with MobileNetV2 for damage detection in slab structures. Fracture and Structural Integrity, 19(72), 121-136. https://doi.org/10.3221/IGF-ESIS.72.09
  2. [20]Ba Anh Le*, Hoang Quan Nguyen, Bao Viet Tran, Thai Son Vu, Thi Loan Bui (2025). Symbolic regression model for predicting compression strength of steel fiber-reinforced concrete. Transportation Research Procedia, 85, 94-101 (SDCAT 2023). https://doi.org/10.1016/j.trpro.2025.03.138

2024

  1. [19]Ba-Anh Le*, Thai-Son Vu, Hoang-Quan Nguyen, Viet-Hung Vu (2024). A Genetic Programming-Assisted Analytical Formula for Predicting the Permeability of Pervious Concrete. Engineering, Technology & Applied Science Research, 14(3), 14775-14780. https://doi.org/10.48084/etasr.7619
  2. [18]Nguyen, Q. B.*, Nguyen, H. H., & Ha, M. H. (2024). An efficient framework for optimization of nonlinear steel trusses with continuous variables using LightGBM. International Journal of GEOMATE, 26(116), 85-92. https://doi.org/10.21660/2024.116.4398
  3. [17]Fan, S., Nguyen, T. N. H., Ren, H., & Wang, P.* (2024). Data-driven shear strength predictions of prestressed concrete hollow-core slabs. Journal of Building Engineering, 95, 110343. https://doi.org/10.1016/j.jobe.2024.110343
  4. [16]Hoang-Quan Nguyen, Ba-Anh Le*, Bao-Viet Tran, Thai-Son Vu, Thi-Loan Bui (2024). Deep artificial neural network-powered phase field model for predicting damage characteristic in brittle composite under varying configurations. Machine Learning: Science and Technology, 5, 025062. https://doi.org/10.1088/2632-2153/ad52e8
  5. [15]Ba-Anh Le, Bao-Viet Tran*, Thai-Son Vu, Viet-Hung Vu, Van-Hung Nguyen (2024). Predicting the Compressive Strength of Pervious Cement Concrete based on Fast Genetic Programming Method. Arabian Journal for Science and Engineering, 49(4), 5487–5504. https://doi.org/10.1007/s13369-023-08396-2

2023

  1. [14]Nguyen, H. D., & Abdel Wahab, M.* (2023). Damage detection in slab structures based on two-dimensional curvature mode shape method and Faster R-CNN. Advances in Engineering Software, 176, 103371. https://doi.org/10.1016/j.advengsoft.2022.103371
  2. [13]Dao, H., Khuc, T.*, Truong, Q., Dinh, C., & Nguyen, A. (2023). Pavement crack segmentation using an attention-based deep learning model. Lecture Notes in Civil Engineering, 727-737. Springer, Singapore. https://doi.org/10.1007/978-981-99-2345-8_75
  3. [12]Nguyen, T. H.*, Nguyen, H. D., & Vu, A. T. (2023). Solving Engineering Optimization Problems Using Machine Learning Classification-Assisted Differential Evolution. Proceedings of the International Conference of Steel and Composite for Engineering Structures. Lecture Notes in Civil Engineering, Springer, 1-23. https://doi.org/10.1007/978-3-031-24041-6_1

2022

  1. [11]Lê Bá Anh, Vũ Thái Sơn, Nguyễn Văn Hưng, Nguyễn Thị Hương Giang, Trần Bảo Việt (2022). Kết hợp phát triển giải tích và kỹ thuật trí tuệ nhân tạo để xây dựng phương trình thấm Kozeny-Carman cho vật liệu bê tông xi măng rỗng. Hội nghị Cơ học toàn quốc lần thứ XI, Tập 2, 470-476.
  2. [10]Ba-Anh Le, Viet-Hung Vu, Soo-Yeon Seo, Bao-Viet Tran*, Tuan Nguyen-Sy, Minh-Cuong Le, Thai-Son Vu (2022). Predicting the Compressive Strength and the Effective Porosity of Pervious Concrete Using Machine Learning Methods. KSCE Journal of Civil Engineering, 26, 4664–4679. https://doi.org/10.1007/s12205-022-1918-z
  3. [9]Vũ Thái Sơn, Vũ Việt Hưng, Nguyễn Tuấn Cường, Trương Đình Thảo Anh, Trần Bảo Việt (2022). Xây dựng công thức dự báo hệ số thấm của bê tông rỗng dựa trên định hướng dữ liệu. Tạp chí Khoa học Giao thông Vận tải, 73(2), 176-188. https://doi.org/10.47869/tcsj.73.2.7

2021

  1. [8]Ho, L. V., Nguyen, D. H., Mousavi, M., De Roeck, G., Bui-Tien, T., Gandomi, A. H., & Wahab, M. A. (2021). A hybrid computational intelligence approach for structural damage detection using marine predator algorithm and feedforward neural networks. Computers & Structures, 252, 106568.
  2. [7]Ho, L. V., Nguyen, H. D., De Roeck, G., Bui-Tien, T., & Abdel Wahab, M.* (2021). Damage detection in steel plates using feed-forward neural network coupled with hybrid particle swarm optimization and gravitational search algorithm. Journal of Zhejiang University-SCIENCE A, 22(6), 467-480. https://doi.org/10.1631/jzus.A2000316
  3. [6]Nguyen, D. H., Bui-Tien, T., Roeck, G. D., & Wahab, M. A. (2021). Damage detection in structures using modal curvatures gapped smoothing method and deep learning. Structural Engineering and Mechanics, 77(1), 47-56.

2020

  1. [5]Nguyen, T. T., Pham Duy, H., Pham Thanh, T., & Vu, H. H. (2020). Compressive Strength Evaluation of Fiber-Reinforced High-Strength Self-Compacting Concrete with Artificial Intelligence. Advances in Civil Engineering, 2020.
  2. [4]Nguyen, D. H., Nguyen, Q. B., Bui-Tien, T., De Roeck, G., & Wahab, M. A. (2020). Damage detection in girder bridges using modal curvatures gapped smoothing method and Convolutional Neural Network: Application to Bo Nghi bridge. Theoretical and Applied Fracture Mechanics, 109, 102728.
  3. [3]Nguyen, D. H., Tran-Ngoc, H., Bui-Tien, T., De Roeck, G., & Wahab, M. A. (2020). Damage detection in truss bridges using transmissibility and machine learning algorithm: Application to Nam O bridge. Smart Structures and Systems, 26(1), 35-47.

2019

  1. [2]Nguyen H.D., Bui T.T., De Roeck G., Abdel Wahab M. (2019) Damage Detection in Ca-Non Bridge Using Transmissibility and Artificial Neural Networks. Structural Engineering and Mechanics. An International Journal
  2. [1]Nguyen H.D., Bui T.T., De Roeck G., Abdel Wahab M. (2019) Damage Detection in Simply Supported Beam Using Transmissibility and Auto-Associative Neural Network. In: Abdel Wahab M. (eds) . Proceedings of the 1st International Conference on Numerical Modelling in Engineering. NME 2018.

How the list was made: every publication on the profiles of current and retired lecturers was assigned to one or more of the 8 areas from its title and venue; a publication can appear in several areas.