Deep learning architecture to predict natural vibration frequencies of damaged structures
Email:
sang.tothanh@uah.edu.vn
Từ khóa:
DNN, Natural Frequency, Surrogate Model, Structural Health Monitoring, SHM
Tóm tắt
Structural damage can arise from various unforeseen causes. Such damage exerts a substantial impact on the load-carrying capacity of the structure. In this study, we propose a Deep Neural Network (DNN) serving as a surrogate model to determine the severity of damage in beam structures Initially, a finite element model (FEM) was constructed in MATLAB to generate the training and testing datasets. Subsequently, a multi-layer deep learning architecture utilizing an artificial neural network is constructed. The Deep Neural Network is trained on this dataset, which encompasses numerous damage scenarios, to predict the output parameters (specifically, the first three natural frequencies of the structure). The reliability of the Deep Neural Network was subsequently verified on the test dataset, achieving an R^2value greater than 0.99. Consequently, this Deep Neural Network can serve as a substitute for the finite element method, thereby significantly accelerating the model updating process within the damage prediction frameworkTài liệu tham khảo
[1]. T. Sang-To, S. Khatir, H. Le-Minh and T. Cuong-Le, Damage detection of steel roof systems using planet optimization algorithm, Journal of Soft Computing in Civil Engineering, 9 (2025) 78-94. https://doi.org/10.22115/scce.2024.452343.1844
[2]. H. Minh, T. Sang-To, B. Le-Van and T. Cuong-Le, Optimization of high-rise concrete structure using a sequence design based on a proposed surrogate model (PSM) and finite element (FE) model updating, Archives of Civil and Mechanical Engineering, 25 (2025) 243. https://doi.org/10.1007/s43452-025-01296-2
[3]. N. Tran, H. Nguyen, T. Nguyen, T. Bui and L. Wang, An advanced hybrid deep learning framework for structural health monitoring of cable-stayed bridge model using fiber optic sensors, Structures, 82 (2025) 110596. https://doi.org/10.1016/j.istruc.2025.110596
[4]. T. S. To, Q. Thien-Huynh, L. Bui and T. Cuong-Le, A Calibrated Framework Integrating Three-Dimensional Numerical Modeling with an Optimization Algorithm for the Estimation of Soil–Pile Interaction,Transportation Infrastructure Geotechnology, 13 (2026) 36. https://doi.org/10.1007/s40515-025-00783-6
[5]. T. Sang-To, T. Tran and T. Cuong-Le, Investigation of the Effectiveness of Optimization Algorithms in Structures, 747 (2025) 241-248. https://doi.org/10.1007/978-3-032-04645-1_29
[6]. L. YiFei, C. MaoSen, T. N. Hoa, S. Khatir, H.-L. Minh, T. SangTo, T. Cuong-Le and M. A. Wahab, Metamodel-assisted hybrid optimization strategy for model updating using vibration response data, Advances in Engineering Software, 185 (2023) 103515. https://doi.org/10.1016/j.advengsoft.2023.103515
[7]. T. To, H. Minh, T. Huynh, S. Khatir, M. Wahab and T. Cuong‐Le, A nonlinear optimization method for calibration of large‐scale deep cement mixing in very soft clay deep excavation, International Journal for Numerical and Analytical Methods in Geomechanics, 48 (2024) 1949-1978. https://doi.org/10.1002/nag.3714
[8]. H. The, T. Tien and H. Ngoc, Deep Learning‐Based Missing Data Reconstruction in SHM Using Gated Dilated Convolution and GRU, International Journal for Numerical Methods in Engineering, 127 (2026), e70257. https://doi.org/10.1002/nme.70257
[9]. M. Görtz, C. Brandl, A. Nitschke, A. Riediger, D. Stromer, M. Byczkowski, V. Heuveline and a. M. Weidemüller, Digital twins for personalized treatment in uro-oncology in the era of artificial intelligence, Nature Reviews Urology, 23 (2026) 29-39. https://doi.org/10.1038/s41585-025-01096-6
[10]. E. M. Dhamani N, Introduction to generative AI, I. Simon and Schuster, 2026.
[11]. H. Minh, T. Sang-To, S. Khatir, M. Wahab, A. Gandomi and T. Cuong-Le, Augmented deep neural network architecture for assessing damage severity in 3D concrete buildings under temperature fluctuations based on K-means optimization, Structures, 57 (2023) 105278. https://doi.org/10.1016/j.istruc.2023.105278
[12]. N.H. Xuan, T.V. Manh, S.N. Nam, L.N. Ngoc, Vibration-based damage detection in cable-stayed bridges using a novel 1D-ConvNeXt-LSTM network, Transport and Communications Science Journal, 77 (2026) 485-499. https://doi.org/10.47869/tcsj.77.4.11
[13]. Q. Le Kha, T.K. Thien, T.N. Thoi, T.N.. Trang, NV. Ho, Reliability-based multi-objective optimization of laminated plates using GDE3 and isogeometric analysis, Transport and Communications Science Journal, 77 (2026) 357-370. https://doi.org/10.47869/tcsj.77.4.2
[14]. H. Hasani and F. Freddi, Condition-aware AI framework for automated structural health monitoring, Automation in Construction, 183 (2026) 106748. https://doi.org/10.1016/j.autcon.2025.106748
[15]. G. Gomes and V. Takano, Strain-based identification of multiple damages in plate-like structures using artificial intelligence and metaheuristic optimization, Machine Learning for Computational Science and Engineering, 2 (2026) 5. https://doi.org/10.1007/s44379-025-00052-w
[16]. A. Ferreira, MATLAB Codes for Finite Element Analysis, Springer, 2008.
[17]. T. Szandała, Review and Comparison of Commonly Used Activation, Bio-inspired neurocomputing, (2020) 203-224. https://doi.org/10.1007/978-981-15-5495-7_11
[2]. H. Minh, T. Sang-To, B. Le-Van and T. Cuong-Le, Optimization of high-rise concrete structure using a sequence design based on a proposed surrogate model (PSM) and finite element (FE) model updating, Archives of Civil and Mechanical Engineering, 25 (2025) 243. https://doi.org/10.1007/s43452-025-01296-2
[3]. N. Tran, H. Nguyen, T. Nguyen, T. Bui and L. Wang, An advanced hybrid deep learning framework for structural health monitoring of cable-stayed bridge model using fiber optic sensors, Structures, 82 (2025) 110596. https://doi.org/10.1016/j.istruc.2025.110596
[4]. T. S. To, Q. Thien-Huynh, L. Bui and T. Cuong-Le, A Calibrated Framework Integrating Three-Dimensional Numerical Modeling with an Optimization Algorithm for the Estimation of Soil–Pile Interaction,Transportation Infrastructure Geotechnology, 13 (2026) 36. https://doi.org/10.1007/s40515-025-00783-6
[5]. T. Sang-To, T. Tran and T. Cuong-Le, Investigation of the Effectiveness of Optimization Algorithms in Structures, 747 (2025) 241-248. https://doi.org/10.1007/978-3-032-04645-1_29
[6]. L. YiFei, C. MaoSen, T. N. Hoa, S. Khatir, H.-L. Minh, T. SangTo, T. Cuong-Le and M. A. Wahab, Metamodel-assisted hybrid optimization strategy for model updating using vibration response data, Advances in Engineering Software, 185 (2023) 103515. https://doi.org/10.1016/j.advengsoft.2023.103515
[7]. T. To, H. Minh, T. Huynh, S. Khatir, M. Wahab and T. Cuong‐Le, A nonlinear optimization method for calibration of large‐scale deep cement mixing in very soft clay deep excavation, International Journal for Numerical and Analytical Methods in Geomechanics, 48 (2024) 1949-1978. https://doi.org/10.1002/nag.3714
[8]. H. The, T. Tien and H. Ngoc, Deep Learning‐Based Missing Data Reconstruction in SHM Using Gated Dilated Convolution and GRU, International Journal for Numerical Methods in Engineering, 127 (2026), e70257. https://doi.org/10.1002/nme.70257
[9]. M. Görtz, C. Brandl, A. Nitschke, A. Riediger, D. Stromer, M. Byczkowski, V. Heuveline and a. M. Weidemüller, Digital twins for personalized treatment in uro-oncology in the era of artificial intelligence, Nature Reviews Urology, 23 (2026) 29-39. https://doi.org/10.1038/s41585-025-01096-6
[10]. E. M. Dhamani N, Introduction to generative AI, I. Simon and Schuster, 2026.
[11]. H. Minh, T. Sang-To, S. Khatir, M. Wahab, A. Gandomi and T. Cuong-Le, Augmented deep neural network architecture for assessing damage severity in 3D concrete buildings under temperature fluctuations based on K-means optimization, Structures, 57 (2023) 105278. https://doi.org/10.1016/j.istruc.2023.105278
[12]. N.H. Xuan, T.V. Manh, S.N. Nam, L.N. Ngoc, Vibration-based damage detection in cable-stayed bridges using a novel 1D-ConvNeXt-LSTM network, Transport and Communications Science Journal, 77 (2026) 485-499. https://doi.org/10.47869/tcsj.77.4.11
[13]. Q. Le Kha, T.K. Thien, T.N. Thoi, T.N.. Trang, NV. Ho, Reliability-based multi-objective optimization of laminated plates using GDE3 and isogeometric analysis, Transport and Communications Science Journal, 77 (2026) 357-370. https://doi.org/10.47869/tcsj.77.4.2
[14]. H. Hasani and F. Freddi, Condition-aware AI framework for automated structural health monitoring, Automation in Construction, 183 (2026) 106748. https://doi.org/10.1016/j.autcon.2025.106748
[15]. G. Gomes and V. Takano, Strain-based identification of multiple damages in plate-like structures using artificial intelligence and metaheuristic optimization, Machine Learning for Computational Science and Engineering, 2 (2026) 5. https://doi.org/10.1007/s44379-025-00052-w
[16]. A. Ferreira, MATLAB Codes for Finite Element Analysis, Springer, 2008.
[17]. T. Szandała, Review and Comparison of Commonly Used Activation, Bio-inspired neurocomputing, (2020) 203-224. https://doi.org/10.1007/978-981-15-5495-7_11
Tải xuống
Chưa có dữ liệu thống kê
Nhận bài
04/03/2026
Nhận bài sửa
04/06/2026
Chấp nhận đăng
03/07/2026
Xuất bản
15/09/2026
Chuyên mục
Công trình khoa học
Kiểu trích dẫn
Thanh Sang, T., & Hieu Nguyen, V. (1789405200). Deep learning architecture to predict natural vibration frequencies of damaged structures. Tạp Chí Khoa Học Giao Thông Vận Tải, 77(7), 944-954. https://doi.org/10.47869/tcsj.77.7.5
