The anomaly detection and health monitoring of dams have attracted increasing attention. To detect the temporal and spatial anomalies of the dam, a novel spatial–temporal variational autoencoder is proposed. The proposed model is based on the sequential variational autoencoder, and its backbone is fulfilled by the recurrent neural network and graph convolutional network to capture the temporal and spatial features in both the generative and inference models. To obtain a normal pattern, we made an assumption that the normal values should be temporally smooth and spatially similar. Then, the smoothness and similarity-inducing operations are used in the framework of the proposed model. Through the addition of smoothness and similarity losses in sequential variational autoencoder, the proposed model can obtain a temporally smooth and spatially similar pattern. For verification, an arch dam is taken as an example. Through comparison among six baseline models, the proposed model detects the temporal and spatial anomalies accurately and stably.
ChenBHuTHuangZ, et al.A spatio-temporal clustering and diagnosis method for concrete arch dams using deformation monitoring data. Struct Heal Monit2018; 18: 1355–1371.
2.
RenQLiMLiH, et al.A robust prediction model for displacement of concrete dams subjected to irregular water-level fluctuations. Computer-Aided Civil Infrastruct Eng2021; 36(5): 577–601.
3.
ShuXBaoTLiY, et al.Dam safety evaluation based on interval-valued intuitionistic fuzzy sets and evidence theory. Sensors2020; 20: 2648.
4.
LiYBaoTGongJ, et al.The prediction of dam displacement time series using STL, extra-trees, and stacked LSTM neural network. IEEE Access2020; 8: 99.
5.
ChenSGuCLinC, et al.Prediction of arch dam deformation via correlated multi-target stacking. Appl Math Model2021; 91: 1175–1193.
6.
DiegoFDavidMOscarM, et al.Automatic bearing fault diagnosis based on one-class v-SVM. Comput Ind Eng2013; 64(1): 357–365.
7.
MontaguePKimJ. An efficient semi-supervised SVM for anomaly detection. 2017 international joint conference on neural networks, Anchorage, AK, USA, 14–19 May 2017, 2017, pp. 2843–2850.
8.
LesoupleJBaudoinCSpigaiM, et al.How to introduce expert feedback in one-class support vector machines for anomaly detection?Signal Process2021; 188(3): 108197.
9.
SalazarFCondeAIrazábalJ, et al.Anomaly detection in dam behaviour with machine learning classification models. Water2021; 13(17): 2387.
10.
FisherWDCampTKKrzhizhanovskayaVV. Anomaly detection in earth dam and levee passive seismic data using support vector machines and automatic feature selection. J Comput Sci2017; 20: 143–153.
11.
ZhangY-MWangHWanH-P, et al.Anomaly detection of structural health monitoring data using the maximum likelihood estimation-based Bayesian dynamic linear model. Struct Health Monit2021; 20(6): 2936–2952.
12.
NguyenLHGouletJ-A. Real-time anomaly detection with Bayesian dynamic linear models. Struct Control Health Monit2019; 26: e2404.
13.
WangXDuYLinS, et al.adVAE: a self-adversarial variational autoencoder with Gaussian anomaly prior knowledge for anomaly detection. Knowledge-Based Syst2019; 190: 190.
14.
MalekiSMalekiSJenningsNR. Unsupervised anomaly detection with LSTM autoencoders using statistical data-filtering. Appl Soft Comput2021; 108(2): 107443.
15.
ShuXBaoTXuR, et al.Dam anomaly assessment based on sequential variational autoencoder and evidence theory. Appl Math Model2021; 98: 576–594.
16.
ChenHChengHJiangG, et al.Exploiting local and global invariants for the management of large scale information systems. In: Eighth IEEE international conference on data mining. IEEE, Pisa, Italy, 15–19 December 2008, 2008, 113–122.
17.
ZhaoSRenHYuanA, et al.Bias and generalization in deep generative models: an empirical study. Proc Adv Neural Inf Process Syst2018; 31: 10792–10801.
18.
LiLYanJWangH, et al. Anomaly detection of time series with smoothness-inducing sequential variational auto-encoder. IEEE Trans Neural Networks Learn Syst2020; 99: 1–15.
19.
MalhotraPRamakrishnanAAnandG, et al.LSTM-based encoder-decoder for multi-sensor anomaly detection. arXiv 2016, arXiv: 1607.00148.
20.
ZhouXShenFLiuL, et al.Graph convolutional network hashing. IEEE Trans Cybernetics2018; 50: 1460–1472.
21.
SimonovskyMKomodakisN. Dynamic edge-conditioned filters in convolutional neural networks on graphs. In: 2017 IEEE conference on computer vision and pattern recognition (CVPR), Hawii, July. IEEE, 2017, pp. 29–38.
22.
XieRJanNMHaoK, et al.Supervised variational autoencoders for soft sensor modeling with missing data. IEEE Trans Ind Inform2020; 16(4): 2820–2828.
23.
MaXLinYNieZ, et al.Structural damage identification based on unsupervised feature-extraction via variational auto-encoder. Measurement2020; 160: 107811.
24.
BleiDMKucukelbirAMcauliffeJD, et al.Variational inference: a review for statisticians. J Am Stat Assoc2017; 112(518): 859–877.
25.
XieRJanNMHaoK, et al.Supervised variational autoencoders for soft sensor modeling with missing data. IEEE Trans Ind Inform2020; 16(4): 2820–2828.
26.
ZhangCSongDChenY, et al.A deep neural network for unsupervised anomaly detection and diagnosis in multivariate time series data. Proc 33rd AAAI Conf. Artif Intell2019; 33: 1409–1416.
27.
GaleanoPPeñaDTsayRS. Outlier detection in multivariate time series by projection pursuit. J Amer Stat Assoc2006; 101: 654–669.
28.
KieuTYangBJensenCS. Outlier detection for multidimensional time series using deep neural networks. In: 19th IEEE international conference on mobile data management (MDM), Aalborg, June. IEEE, 2018, pp. 125–134.
29.
ZhuDMaYLiuY. Anomaly detection with deep graph autoencoders on attributed networks. In: 2020 IEEE symposium on computers and communications (ISCC), Rennes, July. IEEE, 2020, pp. 1–6.
30.
DinhPTParkM and R-EDoS: robust economic denial of sustainability detection in an SDN-based cloud through stochastic recurrent neural network. IEEE Access2021; 9: 35057–35074.
31.
DavisJ. The relationship between precision-recall and ROC curves. Proceedings 23th international conference on machine learn, Pennsylvania, June, 2006, pp. 233–240.
32.
CongTTanROttewillJR, et al.Anomaly detection and mode identification in multimode processes using the field Kalman filter. IEEE Trans Control Syst Technol2021; 29(5): 2192–2205.
