Van-Phuong Ha1This email address is being protected from spambots. You need JavaScript enabled to view it., Dinh-Van Nguyen2, Trong-Chuong Trinh1, Duc-Cuong Quach1, and Van HuyBui1
1Hanoi University of Industry, 298 Cau Dien, Bac Tu Liem, Hanoi, Vietnam
2Hanoi University of Science and Technology, School of Electrical and Electronic Engineering, 01 Dai Co Viet, Hai Ba Trung, Hanoi, Vietnam
Received: April 8, 2024 Accepted: September 19, 2024 Publication Date: November 30, 2024
Copyright The Author(s). This is an open access article distributed under the terms of the Creative Commons Attribution License (CC BY 4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are cited.
Most modern hydroelectric dams are equipped with state-of-the-art sensors to measure and detect irregularities as part of a comprehensive safety-monitoring system. To better prevent future disasters, machine-learning algorithms have been employed. Often, these algorithms are trained on historical sensor data to predict future events. However, owing to dams’ high safety standards, it is uncommon for significant anomalies to occur. Hence, most of the collected sensor measurements were skewed to the normal range. This often results in the collection of extremely imbalanced datasets. Furthermore, the sensor measurements are highly correlated with each other along the spatial and temporal axes. Thus, the sensor measurement input vector for machine learning algorithms should be a multidimensional vector instead of a one-dimensional vector, as suggested by the traditional tabular form of data. In this study, a novel method for addressing these problems is proposed. By converting the collected sensor data into images, it is possible to include both the temporal and spatial relationships of the data in the input vector. Moreover, an autoencoder was built to detect anomalies in an imbalanced dataset. To validate the proposed method, nearly seven years of sensor data collected from a critical section of a dam structure were used. The data are then grouped along the temporal axis and converted into an image that represents a single day of the sensor measurements. Using these images along with the proposed autoencoder structure, the method achieves an AUC score of 0.97 in detecting anomalies given a single day of sensor image.
Keywords: Sensor; Hydroelectric dam safety; Deep learning; Autoencoder; Anomaly detection; Imbalanced dataset
[1] T. P. Carvalho, F. A. Soares, R. Vita, R. d. P. Fran cisco, J. P. Basto, and S. G. Alcalá, (2019) “A systematic literature review of machine learning methods applied to predictive maintenance" Computers & Industrial Engi neering 137: 106024. DOI: 10.1016/j.cie.2019.106024.
[2] B. Chen, T. Hu, Z. Huang, and C. Fang, (2019) “A spatio-temporal clustering and diagnosis method for con crete arch dams using deformation monitoring data" Structural Health Monitoring 18(5-6): 1355–1371. DOI: 10.1177/1475921718797949.
[3] T. Chen, Q. Duan, X. Zheng, R. Gan, and M. Pan, (2020) “Dam deformation prediction based on eemd sarima mode" The International Archives of the Pho togrammetry, Remote Sensing and Spatial Infor mation Sciences 42: 1091–1097. DOI: 10.5194/isprs archives-XLII-3-W10-1091-2020.
[4] F. Dama and C. Sinoquet, (2021) “Time series analy sis and modeling to forecast: A survey" arXiv preprint arXiv:2104.00164:
[5] A.Gaagai, H.A.Aouissi, A. E. Krauklis, J. Burlakovs, A. Athamena, I. Zekker, A. Boudoukha, L. Be naabidate, and H. Chenchouni, (2022) “Modeling and risk analysis of dam-break flooding in a semi-arid Montane watershed: a case study of the Yabous Dam, Northeastern Algeria" Water 14(5): 767. DOI: 10.3390/w14050767.
[6] GEOKON.Crackmeters (VW) | Model 4420. 2024.
[7] S. Jeong, M. Ferguson, R. Hou, J. P. Lynch, H. Sohn, andK.H.Law,(2019)“Sensordatareconstruction using bidirectional recurrent neural network with application to bridge monitoring" Advanced Engineering Informat ics 42: 100991. DOI: https://doi.org/10.1016/j.aei.2019.100991.
[8] B. T. Kieu Trinh, X. Yangxuan, C. Van Doan, D. Xuan Khanh, T. The Viet, and M. Dinh Sinh, (2021) “Ap plication of Statistic Model and Backpropagation Neural Network to Analyzing and Forecasting Hydropower Dam Displacement" VNU Journal of Science: Earth and Environmental Sciences 37(1): DOI: 10.25073/2588-1094/vnuees.4529.
[9] J. Kong, W. Kowalczyk, S. Menzel, and T. Bäck. “Im proving imbalanced classification by anomaly detec tion”. In: Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lec ture Notes in Bioinformatics). 12269 LNCS. Springer Sci ence and Business Media Deutschland GmbH, 2020, 512–523. DOI: 10.1007/978-3-030-58112-1_35.
[10] B. Li, J. Yang, and D. Hu, (2020) “Dam monitoring data analysis methods: A literature review" Structural Control and Health Monitoring 27(3): DOI: 10.1002/stc.2501.
[11] C. Lin, S. Chen, M. A. Hariri-Ardebili, and T. Li, (2023) “An Explainable Probabilistic Model for Health Monitoring of Concrete Dam via Optimized Sparse Bayesian Learning and Sensitivity Analysis" Structural Control and Health Monitoring 2023: DOI: 10.1155/2023/2979822.
[12] X. Liu, Z. Li, L. Sun, E. Y. Khailah, J. Wang, and W. Lu, (2023) “A critical review of statistical model of dam monitoring data" Journal of Building Engineering 80: 108106. DOI: https://doi.org/10.1016/j.jobe.2023.108106.
[13] L. S. Ribeiro, V. E. Wilhelm, É. F. Faria, J. M. Cor rea, and A. C. P. dos Santos, (2019) “A comparative analysis of long-term concrete deformation models of a but tress dam" Engineering Structures 193: 301–307. DOI: https://doi.org/10.1016/j.engstruct.2019.05.043.
[14] J. Rico, J. Barateiro, J. Mata, A. Antunes, and E. Car doso, (2019) “Applying advanced data analytics and ma chine learning to enhance the safety control of dams" Ma chine learning paradigms: Applications of learning and analytics in intelligent systems: 315–350. DOI: 10.1007/978-3-030-15628-2_10.
[15] F. Salazar, A. Conde, J. Irazábal, and D. J. Vicente, (2021) “Anomaly detection in dam behaviour with ma chine learning classification models" Water (Switzer land) 13(17): DOI: 10.3390/w13172387.
[16] T. Khanh and L. Tinh, (2010) “Applying a method of correlation analysis to estimate the movement of construc tion" Journal of Mining and Geology 4(30):
[17] D. Valle, J. Hyde, M. Marsik, and S. Perz, (2020) “Im proved inference and prediction for imbalanced binary big data using case-control sampling: A case study on defor estation in the Amazon region" Remote Sensing 12(8): DOI: 10.3390/RS12081268.
We use cookies on this website to personalize content to improve your user experience and analyze our traffic. By using this site you agree to its use of cookies.
