Multi-Resolution Wavelet-Graph Fusion for Early Fatigue Crack Detection in Steel Structures with Limited Labeled Data
DOI:
https://doi.org/10.54097/95pr0369Keywords:
Fatigue crack detection, wavelet transform, graph convolutional network, structural health monitoring, semi-supervised learning, relative wavelet packet entropy, steel structuresAbstract
Early detection of fatigue cracks in steel structures remains a fundamental challenge in structural health monitoring (SHM), particularly under conditions where labeled training data are scarce. This paper proposes a multi-resolution wavelet-graph fusion framework that integrates continuous wavelet transform (CWT) with graph convolutional networks (GCN) to simultaneously capture time-frequency signal characteristics and spatial inter-sensor dependencies. A semi-supervised learning (SSL) strategy exploits unlabeled vibration measurements to supplement limited annotated samples. Wavelet scalogram features extracted at three resolution levels are encoded onto a sensor graph derived from a five-story shear-building benchmark model, where graph topology reflects physical structural connectivity. Relative wavelet packet entropy (RWPE) computed across nine sensor locations serves as the primary damage-sensitive feature, revealing spatially localized crack signatures that single-channel methods fail to detect. A four-phase experimental framework processes multi-sensor acceleration data through normalization, wavelet transform, and hierarchical nonlinear principal component analysis (h-NLPCA) to classify seven structural states. Experiments demonstrate 93.7% detection accuracy using only 15% labeled samples, outperforming baselines by up to 11.2 percentage points. The results confirm that wavelet-graph fusion provides a robust and data-efficient solution for real-world steel infrastructure inspection.
Downloads
References
[1] Liu, C. L., Tseng, C. J., Huang, T. H., Yang, J. S., & Huang, K. B. (2023). A multi-task learning model for building electrical load prediction. Energy and Buildings, 278, 112601. https://doi.org/10.1016/j.enbuild.2022.112601
[2] Chen, J., Liang, Y., Liu, J., & Zhou, M. (2026). Temporal Transformer with Conditional Tabular GAN for Credit Card Fraud Detection: A Sequential Deep Learning Approach. Mathematics, 14(7), 1183. https://doi.org/10.3390/math14071183
[3] Wang, B., Wang, Z., Zhao, W., Zhang, F., & Shang, W. (2026). DRL-Adapt: Deep Reinforcement Learning for Adaptive Routing Convergence Optimization in Large-Scale Networks. IEEE Open Journal of the Computer Society. https://doi.org/10.1109/OJCS.2026.xxxxxx
[4] Teng, D., Rhee, M., Qin, Y., Zi, B., & Liu, W. (2026). SW-SpeedDLM: Sliding-Window Speculative Decoding for Diffusion Language Models under Long-Context Constraints. Mathematics.
[5] Avci, O., Abdeljaber, O., Kiranyaz, S., Sassi, S., Ibrahim, A., & Gabbouj, M. (2021). One-dimensional convolutional neural networks for real-time damage detection of rotating machinery. In Rotating Machinery, Optical Methods & Scanning LDV Methods, Volume 6: Proceedings of the 39th IMAC, A Conference and Exposition on Structural Dynamics 2021 (pp. 73–83). Springer International Publishing. https://doi.org/10.1007/978-3-030-76232-6_9
[6] Bao, Y., Tang, Z., Li, H., & Zhang, Y. (2019). Computer vision and deep learning–based data anomaly detection method for structural health monitoring. Structural Health Monitoring, 18(2), 401–421. https://doi.org/10.1177/1475921718763241
[7] Tang, Z., Chen, Z., Bao, Y., & Li, H. (2019). Convolutional neural network‐based data anomaly detection method using multiple information for structural health monitoring. Structural Control and Health Monitoring, 26(1), e2296. https://doi.org/10.1002/stc.2296
[8] Rosafalco, L., Manzoni, A., Mariani, S., & Corigliano, A. (2020). Fully convolutional networks for structural health monitoring through multivariate time series classification. Advanced Modeling and Simulation in Engineering Sciences, 7(1), 38. https://doi.org/10.1186/s40323-020-00135-9
[9] Ghiasi, A., Ng, C. T., & Sheikh, A. H. (2022). Damage detection of in-service steel railway bridges using a fine k-nearest neighbor machine learning classifier. Structures, 45, 1920–1935. https://doi.org/10.1016/j.istruc.2022.09.095
[10] D’Angela, D., & Ercolino, M. (2021). Fatigue crack growth in metallic components: Numerical modelling and analytical solution. Structural Engineering and Mechanics, 79(5), 541–556. https://doi.org/10.12989/sem.2021.79.5.541
[11] Wei, P., Li, Q., Sun, M., & Huang, J. (2023). Modal identification of high-rise buildings by combined scheme of improved empirical wavelet transform and Hilbert transform techniques. Journal of Building Engineering, 63, 105443. https://doi.org/10.1016/j.jobe.2023.105443
[12] Islam, M. M., & Kim, J. M. (2019). Automated bearing fault diagnosis scheme using 2D representation of wavelet packet transform and deep convolutional neural network. Computers in Industry, 106, 142–153. https://doi.org/10.1016/j.compind.2019.01.004
[13] Rajendran, M., & Subbian, D. (2026). Deep learning in corrosion assessment and control: a critical review of techniques and challenges. Corrosion Reviews, 44(1), 20240060. https://doi.org/10.1515/corrrev-2024-0060
[14] Liu, T., Xu, H., Ragulskis, M., Cao, M., & Ostachowicz, W. (2020). A data-driven damage identification framework based on transmissibility function datasets and one-dimensional convolutional neural networks: Verification on a structural health monitoring benchmark structure. Sensors, 20(4), 1059. https://doi.org/10.3390/s20041059
[15] Bloemheuvel, S., van den Hoogen, J., & Atzmueller, M. (2021). A computational framework for modeling complex sensor network data using graph signal processing and graph neural networks in structural health monitoring. Applied Network Science, 6(1), 97. https://doi.org/10.1007/s41109-021-00339-7
[16] Khodabandehlou, H., Pekcan, G., & Fadali, M. S. (2019). Vibration‐based structural condition assessment using convolution neural networks. Structural Control and Health Monitoring, 26(2), e2308. https://doi.org/10.1002/stc.2308
[17] Sony, S., Gamage, S., Sadhu, A., & Samarabandu, J. (2022). Multiclass damage identification in a full-scale bridge using optimally tuned one-dimensional convolutional neural network. Journal of Computing in Civil Engineering, 36(2), 04021035. https://doi.org/10.1061/(ASCE)CP.1943-5487.0001035
[18] Chaabane, M., Mansouri, M., Ben Hamida, A., Nounou, H., & Nounou, M. (2019). Multivariate statistical process control‐based hypothesis testing for damage detection in structural health monitoring systems. Structural Control and Health Monitoring, 26(1), e2287. https://doi.org/10.1002/stc.2287
[19] Zhang, S., Qiu, L., & Zeng, Z. (2026). Physics-Data Synergy in Structural Health Monitoring: A Multi-Scale Graph Contrastive Framework With Temperature-Adaptive Fusion. IEEE Access. https://doi.org/10.1109/ACCESS.2026.xxxxxx
[20] Piechnicki, F., Dos Santos, C. F., De Freitas Rocha Loures, E., & Dos Santos, E. A. P. (2021). Data fusion framework for decision-making support in reliability-centered maintenance. Journal of Industrial and Production Engineering, 38(1), 1–17. https://doi.org/10.1080/21681015.2020.1859722
[21] Tiboni, M., Remino, C., Bussola, R., & Amici, C. (2022). A review on vibration-based condition monitoring of rotating machinery. Applied Sciences, 12(3), 972. https://doi.org/10.3390/app12030972
[22] Kong, Q. Y., Rong, Y., Wang, G. L., Xu, Z. Q., Zhang, Q., Guan, Z. S., & Fan, Y. H. (2025). Adaptive Graph Neural Networks with Semi-Supervised Multi-Modal Fusion for Few-Shot Steel Strip Defect Detection. Processes, 13(11), 3520. https://doi.org/10.3390/pr13113520
[23] Noël, J. P., & Schoukens, M. (2019). Cross-fertilising research in nonlinear system identification between the mechanical, control and machine learning fields: Editorial statement. Mechanical Systems and Signal Processing, 130, 213–220. https://doi.org/10.1016/j.ymssp.2019.05.035
[24] Sarmadi, H., Entezami, A., & Daneshvar Khorram, M. (2020). Energy-based damage localization under ambient vibration and non-stationary signals by ensemble empirical mode decomposition and Mahalanobis-squared distance. Journal of Vibration and Control, 26(11–12), 1012–1027. https://doi.org/10.1177/1077546319898229
[25] Mousavi, M., & Gandomi, A. H. (2021). Prediction error of Johansen cointegration residuals for structural health monitoring. Mechanical Systems and Signal Processing, 160, 107847. https://doi.org/10.1016/j.ymssp.2021.107847
[26] Ping, W., Jiao, Y., Fan, H., & Zhang, X. (2026). Multimodal Fraud Detection in Financial Statements: A Trimodal Attention Network with Contrastive Evidence Chain Construction. IEEE Access. https://doi.org/10.1109/ACCESS.2026.xxxxxx
[27] Van Engelen, J. E., & Hoos, H. H. (2020). A survey on semi-supervised learning. Machine Learning, 109(2), 373–440. https://doi.org/10.1007/s10994-019-0587-0
Downloads
Published
Issue
Section
License
Copyright (c) 2026 International Journal of Advanced Engineering and Technology Research
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License.
