تشخیص داده‌محور و بدون نظارت آسیب در سازه‌های قابی شکل،با به‌کارگیری مدل‌های شبکه عصبی عمیق

نوع مقاله : مقاله علمی

نویسندگان
امید اللهیاری پرگو 1
غلامرضا نصیرپور 2
اکبر اصغرزاده بناب 2
امیرحسین باباپور 3
1 کارشناس ارشد، گروه مهندسی عمران، دانشکده فنی و مهندسی، دانشگاه خوارزمی، تهران، ایران
2 استادیار، گروه مطالعات علم و فناوری، دانشگاه فرماندهی و ستاد آجا، تهران، ایران
3 دانشجوی کارشناسی ارشد، دانشکده مهندسی مکانیک، دانشگاه سمنان، سمنان، ایران
10.22034/stme.2025.532643.1142
چکیده
تشخیص آسیب، باهدف جلوگیری از خرابی کلی سازه، ترمیم و بازسازی برنامه‌ریزی شده، حوزه تحقیقات مهم سال‌های اخیر بوده‌است. پایش سلامت سازه با رویکرد داده‌محور بوسیله تحلیل پاسخ سازه موضوع اصلی پژوهش‌های سال‌های اخیر است. در این رویکرد با به‌کارگیری ابزار هوش مصنوعی خصوصاً شبکه‌های عصبی مصنوعی، نیاز به پیش‌پردازش پیچیده داده‌های سری زمانی را برطرف شده و نتایج دقیق‌تری نسبت به رویکرد‌های گذشته پایش سلامت سازه و تشخیص آسیب حاصل شده‌ست. در این پژوهش، یک شبکه عصبی عمیق بدون نظارت، (خودرمزگذار پیچشی) باهدف بازسازی داده ورودی و استفاده از خطای بازسازی آن به‌عنوان شاخصه حساس به آسیب ارائه شده‌است. یافته‌ها نشان داده است مدل‌ مورد نظر، با دقت 774/99%، نتایج دقیقی در تشخیص آسیب وضعیت مختلف سازه با وجود آموزش صرفاً بر اساس داده‌های سالم سازه به همراه داشته؛ همچنین مدل موردنظر ازنظر تعداد پارامترهای آموزش و تلاش محاسباتی عملکرد مطلوبی ارائه داده‌است. در انتها جهت اثبات کارایی مدل، نتایج با پژوهش‌های مشابه مقایسه گردیده که حاکی از نتایجی دقیق‌تر از آن‌ها است.
کلیدواژه‌ها
تشخیص آسیب سازه‌ای
پایش سلامت سازه
تشخیص ناهنجاری
شبکه عصبی عمیق
یادگیری بدون نظارت
موضوعات

عنوان مقاله English

Data-driven and unsupervised damage detection in frame structures using deep neural networks

نویسندگان English

Omid Allahyari Pargu 1
Gholamreza Nasirpour 2
Akbar Asgharzadeh-Bonab 2
Amirhossein Babapour 3
1 M.Sc, Department of Civil Engineering, Faculty of Engineering, Kharazmi University, Tehran, Iran
2 Assistant professor, Department of Science and Technology Studies, AJA Command and Staff University, Tehran, Iran
3 M.Sc. Student, Department of Mechanical Engineering, Semnan University, Semnan, Iran
چکیده English

Damage detection, aimed at preventing overall structural failure and enabling planned repair and rehabilitation, has been a significant research focus in recent years. Data-driven structural health monitoring through structural response analysis constitutes the core of contemporary studies. This approach utilises artificial intelligence tools, particularly artificial neural networks, to eliminate the need for complex preprocessing of time-series data and achieve more accurate results compared to traditional structural health monitoring and damage detection methods. In this study, an unsupervised deep neural network (convolutional autoencoder) is proposed to reconstruct input data and employ the reconstruction error as a damage-sensitive feature. The findings demonstrate that the proposed model achieves highly accurate damage detection across various structural conditions, despite being trained solely on healthy structural data. Moreover, the model exhibits favourable performance in terms of the number of training parameters and computational effort. Finally, to validate the effectiveness of the model, the results are compared with those of similar studies, demonstrating superior accuracy.

کلیدواژه‌ها English

Structural Damage Detection
Structural Health Monitoring
Anomaly Detection
Deep Neural Networks
Unsupervised Learning

اصل مقاله

 

[1] A. Asgharzadeh Bonab, H. Kalb Khani, and S. Bijanvand, “A machine learning–based approach for linear and nonlinear damage detection in structures using combined deep temporal and time–frequency features,” Mechanical Engineering, University of Tabriz, Vol. 54, No. 1, pp. 71–80, 2024 (in Persian). doi: 10.22034/jmeut.2024.60099.3368
[2] M. Azimi, A. Dadras Eslamlou, and G. Pekcan, “Data-driven structural health monitoring and damage detection through deep learning: A state-of-the-art review,” Sensors, Vol. 20, No. 10, p. 2778, 2020. doi: 10.3390/s20102778
[3] Z. Yuan, S. Zhu, C. Chang, X. Yuan, Q. Zhang, and W. Zhai, “An unsupervised method based on convolutional variational auto-encoder and anomaly detection algorithms for light rail squat localization,” Construction and Building Materials, Vol. 313, p. 125563, 2021. doi: 10.1016/j.conbuildmat.2021.125563
[4] K. Eltouny and X. Liang, “Large-scale structural health monitoring using composite recurrent neural networks and grid environments,” Computer-Aided Civil and Infrastructure Engineering, Vol. 38, No. 3, pp. 271–287, 2023. doi: 10.1111/mice.12841
[5] M. Z. Sarwar and D. Cantero, “Probabilistic autoencoder-based bridge damage assessment using train-induced responses,” Mechanical Systems and Signal Processing, Vol. 208, p. 111046, 2024. doi: 10.1016/j.ymssp.2024.111046
[6] O. Avci, O. Abdeljaber, S. Kiranyaz, M. Hussein, M. Gabbouj, and D. Inman, “A review of vibration-based damage detection in civil structures: From traditional methods to Machine Learning and Deep Learning applications,” Mechanical Systems and Signal Processing, Vol. 147, p. 107077, 2021. doi: 10.1016/j.ymssp.2020.107077
[7] A. Asgharzadeh Bonab, G. Nasirpour, and A. Allahyari Pargo, “A comparative study and accurate approach for damage detection of frame structures based on deep neural network models,” Mechanical Engineering, University of Tabriz, Vol. 54, No. 1, pp. 71–80, 2025 (in Persian). doi: 10.22034/jmeut.2025.63709.3467
[8] S. Ghazimoghadam and S. A. A. Hosseinzadeh, “A novel unsupervised deep learning approach for vibration-based damage diagnosis using a multi-head self-attention LSTM autoencoder,” Measurement, Vol. 229, p. 114410, 2024. doi: 10.1016/j.measurement.2024.114410
[9] A. Pollastro, G. Testa, A. Bilotta, and R. Prevete, “Semi-supervised detection of structural damage using variational autoencoder and a one-class support vector machine,” IEEE Access, 2023. doi: 10.1109/ACCESS.2023.3291674
[10] A. Anaissi, S. M. Zandavi, B. Suleiman, M. Naji, and A. Braytee, “Multi-objective variational autoencoder: An application for smart infrastructure maintenance,” Applied Intelligence, Vol. 53, No. 10, pp. 12047–12062, 2023. doi: 10.1007/s10489-022-04321-z
[11] D. Terzis, “Monitoring innovation metrics in construction and civil engineering: Trends, drivers and laggards,” Developments in the Built Environment, Vol. 9, 2022. doi: 10.1016/j.dibe.2022.100054
[12] H. Dang, M. Tatipamula, and H. Nguyen, “Cloud-based digital twinning for structural health monitoring using deep learning,” IEEE Trans. Ind. Informatics, Vol. 18, No. 6, pp. 3820–3830, 2021. doi: 10.1109/TII.2021.3064907
[13] E. Figueiredo, G. Park, J. Figueiras, C. Farrar, and K. Worden, “Structural health monitoring algorithm comparisons using standard data sets,” Los Alamos National Lab, 2009. doi: 10.2172/947958
[14] R. Chalapathy and S. Chawla, “Deep learning for anomaly detection: A survey,” arXiv preprint, arXiv:1901.03407, 2019. Available: doi: 10.48550/arXiv.1901.03407
[15] S. Samudra, M. Barbosh, and A. Sadhu, “Machine learning-assisted improved anomaly detection for structural health monitoring,” Sensors, Vol. 23, No. 7, p. 3365, 2023. doi: 10.3390/s23073365
[16] C. Farrar, S. Doebling, and D. Nix, “Vibration-based structural damage identification,” Philos. Trans. R. Soc. A, Vol. 359, No. 1778, pp. 131–149, 2001. doi: 10.1098/rsta.2000.0735
[17] K. O’Shea, “An introduction to convolutional neural networks,” arXiv preprint, arXiv:1511.08458, 2015. Available: https://doi.org/10.48550/arXiv.1511.08458
[18] G. Marcus, “Deep learning: A critical appraisal,” arXiv preprint, arXiv:1801.00631, 2018. Available: doi: 10.48550/arXiv.1801.00631
[19] R. M. Schmidt, “Recurrent neural networks (RNNs): A gentle introduction and overview,” arXiv preprint, arXiv:1912.05911, 2019. Available: doi:10.48550/arXiv.1912.05911
[20] Y. J. Cha, R. Ali, J. Lewis, and O. Büyüköztürk, “Deep learning-based structural health monitoring,” Automation in Construction, Vol. 161, p. 105328, 2024. doi: 10.1016/j.autcon.2024.105328.
[21] D. Kwon, H. Kim, J. Kim, S. C. Suh, I. Kim, and K. Kim, “A survey of deep learning-based network anomaly detection,” Cluster Computing, Vol. 22, pp. 949–961, 2019. doi: 10.1007/s10586-018-3081-1
[22] S. Chen and W. Guo, “Auto-encoders in deep learning — A review with new perspectives,” Mathematics, Vol. 11, No. 8, p. 1777, 2023. doi: 10.3390/math11081777
[23] O. Abdeljaber, O. Avci, S. Kiranyaz, M. Gabbouj, and D. J. Inman, “Real-time vibration-based structural damage detection using one-dimensional convolutional neural networks,” Journal of Sound and Vibration, Vol. 388, pp. 154–170, 2017. doi: 10.1016/j.jsv.2016.08.050
[24] S. Sharma and S. Sen, “One-dimensional convolutional neural network-based damage detection in structural joints,” Journal of Civil Structural Health Monitoring, Vol. 10, No. 5, pp. 1057–1072, 2020. doi: 10.1007/s13349-020-00423-3
[25] Z. Teng, S. Teng, J. Zhang, G. Chen, and F. Cui, “Structural damage detection based on real-time vibration signal and convolutional neural network,” Applied Sciences, Vol. 10, No. 14, p. 4720, 2020. doi: 10.3390/app10144720
[26] K. Berahmand, F. Daneshfar, E. Salehi, Y. Li, and Y. Xu, “Autoencoders and their applications in machine learning: A survey,” Artificial Intelligence Review, Vol. 57, No. 2, p. 28, 2024. doi: 10.1007/s10462-023-10478-w
[27] U. Michelucci, “An introduction to autoencoders,” arXiv preprint, arXiv:2201.03898, 2022. Available: doi: 10.48550/arXiv.2201.03898
[28] A. Santos, E. Figueiredo, M. Silva, C. Sales, and J. Costa, “Machine learning algorithms for damage detection: Kernel-based approaches,” Journal of Sound and Vibration, Vol. 363, pp. 584–599, 2016. doi: 10.1016/j.jsv.2016.05.033
.[29] L. Balsamo, R. Betti, and H. Beigi, “A structural health monitoring strategy using cepstral features,” Journal of Sound and Vibration, Vol. 333, No. 19, pp. 4526–4542, 2014. doi: 10.1016/j.jsv.2014.05.040
[30] K. Kauss, V. Alves, F. Barbosa, and A. Cury, “Semi-supervised structural damage assessment via autoregressive models and evolutionary optimization,” Structures, Vol. 59, p. 105762, 2024. doi: 10.1016/j.structures.2024.105762
.
 
 

  • تاریخ دریافت 14 تیر 1404
  • تاریخ بازنگری 13 مهر 1404
  • تاریخ پذیرش 09 شهریور 1404
  • تاریخ اولین انتشار 09 شهریور 1404
  • تاریخ انتشار 01 بهمن 1404