Enormous data are continuously collected by the structural health monitoring system of civil infrastructures. The structural health monitoring data inevitably involve anomalies caused by sensors, transmission errors, or abnormal structural behaviors. It is important to identify the anomalies and find their origin (e.g. sensor fault or structural damage) to make correct interventions. Moreover, online anomaly identification of the structural health monitoring data is critical for timely structural condition assessment and decision-making. This study proposes an online approach for detecting anomalies of the structural health monitoring data based on the Bayesian dynamic linear model. In particular, Bayesian dynamic linear model, consisting of various components, is implemented to characterize the feature of real-time measurements. Expectation maximization algorithm and Kalman smoother are combined to estimate the Bayesian dynamic linear model parameters and generate log-likelihood functions. The subspace identification method is introduced to overcome the initialization issue of the expectation maximization algorithm. The log-likelihood difference of consecutive time steps is then used to determine thresholds without introducing extra anomaly detectors. The proposed Bayesian dynamic linear model-based approach is first illustrated by the simulation data and then applied to the structural health monitoring data collected from two long-span bridges. The results indicate that the proposed method exhibits good accuracy and high computational efficiency and also allows for reconstructing the strain measurements to replace anomalies.
KarbhariVMAnsariF. Structural health monitoring of civil infrastructure systems. Amsterdam: Elsevier, 2009.
2.
FarrarCRWordenK. An introduction to structural health monitoring. Philos Trans Royal Soc2007; 365: 303–315.
3.
LiHOuJ. The state of the art in structural health monitoring of cable-stayed bridges. J Civ Struct Health Monit2016; 6: 43–67.
4.
CatbasFNSusoyMFrangopolDM. Structural health monitoring and reliability estimation: long span truss bridge application with environmental monitoring data. Eng Struct2008; 30: 2347–2359.
5.
MaoZToddMD. A Bayesian recursive framework for ball-bearing damage classification in rotating machinery. Struct Health Monit2016; 15: 668–684.
6.
KromanisRKripakaranP. Support vector regression for anomaly detection from measurement histories. Adv Eng Inform2013; 27: 486–495.
7.
NiFZhangJNooriMN. Deep learning for data anomaly detection and data compression of a long-span suspension bridge. Compu-aid Civ Infrast Eng2020; 35: 685–700.
8.
NguyenLHGouletJA. Real-time anomaly detection with Bayesian dynamic linear models. Struct Control Health Monit2019; 26: e2404.
9.
HillDJMinskerBS. Anomaly detection in streaming environmental sensor data: a data-driven modeling approach. Environ Modell Softw2010; 25: 1014–1022.
10.
FarrarCRWordenK. Structural health monitoring: a machine learning perspective. Hoboken, NJ: John Wiley & Sons, 2012.
11.
MaoJXWangHFuYG, et al. Automated modal identification using principal component and cluster analysis: application to a long-span cable-stayed bridge. Struct Control Health Monit2019; 26: e2430.
12.
HillDJMinskerBSAmirE. Real-time Bayesian anomaly detection in streaming environmental data. Water Resour Res2009; 45: 2008WR006956.
13.
AminikhanghahiSCookDJ. A survey of methods for time series change point detection. Knowl Inf Syst2017; 51: 339–367.
14.
MaoJWangHSpencerBFJr. Toward data anomaly detection for automated structural health monitoring: exploiting generative adversarial nets and autoencoders. Struct Health Monit. Epub ahead of print 7 June 2020. DOI: https://doi.org/10.1177/1475921720924601.
15.
BaoYTangZLiH, et al. Computer vision and deep learning–based data anomaly detection method for structural health monitoring. Struct Health Monit2019; 18: 401–421.
16.
TangZChenZBaoY, et al. Convolutional neural network-based data anomaly detection method using multiple information for structural health monitoring. Struct Control Health Monit2019; 26: e2296.
17.
AhmadSLavinAPurdyS, et al. Unsupervised real-time anomaly detection for streaming data. Neurocomputing2017; 262: 134–147.
18.
PangJLiuDPengY, et al. Anomaly detection based on uncertainty fusion for univariate monitoring series. Measurement2017; 95: 280–292.
19.
WangHZhangY-MMaoJ-X, et al. A probabilistic approach for short-term prediction of wind gust speed using ensemble learning. J Wind Eng Ind Aerodyn2020; 202: 104198.
20.
LiXLiYLuX, et al. An online anomaly recognition and early warning model for dam safety monitoring data. Struct Health Monit2020; 19: 796–809.
21.
MoghaddassRShengS. An anomaly detection framework for dynamic systems using a Bayesian hierarchical framework. Appl Energy2019; 240: 561–582.
22.
AgostaJMDiuk-WasserCChandrashekarJ, et al. An adaptive anomaly detector for worm detection. In: Proceedings of the 2nd USENIX workshop on tackling computer systems problems with machine learning techniques, Cambridge, MA, April 2007, p. 1–6. Berkeley, CA: USENIX Association. https://dl.acm.org/doi/proceedings/10.5555/1361442
23.
AliMQAl-ShaerEKhanH, et al. Automated anomaly detector adaptation using adaptive threshold tuning. ACM Trans Inf Syst Secur (TISSEC)2013; 15: 1–30.
24.
GouletJA. Bayesian dynamic linear models for structural health monitoring. Struct Control Health Monit2017; 24: e2035.
25.
LipowskyHStaudacherSBauerM, et al. Application of Bayesian forecasting to change detection and prognosis of gas turbine performance. J Eng Gas Turb Power2010; 132: 031602.
26.
WestMHarrisonJ. Bayesian forecasting and dynamic models. New York: Springer Science & Business Media, 2006.
27.
NguyenLHGouletJA. Anomaly detection with the switching Kalman filter for structural health monitoring. Struct Control Health Monit2018; 25: e2136.
28.
ZaidanMAHarrisonRFMillsAR, et al. Bayesian hierarchical models for aerospace gas turbine engine prognostics. Expert Syst Appl2015; 42: 539–553.
29.
WangYNiYWangX. Real-time defect detection of high-speed train wheels by using Bayesian forecasting and dynamic model. Mech Syst Signal Process2020; 139: 106654.
30.
WangHZhangY-MMaoJ-X, et al. Modeling and forecasting of temperature-induced strain of a long-span bridge using an improved Bayesian dynamic linear model. Eng Struct2019; 192: 220–232.
31.
OcakHLoparoKA. HMM-based fault detection and diagnosis scheme for rolling element bearings. J Vib Acoust2005; 127: 299–306.
32.
LiZOuyangYSuL, et al. Detecting traffic anomaly in wireless networks, an analytics methodology. In: Proceedings of the 2018 wireless telecommunications symposium (WTS), 17–20 April 2018, pp. 1–6. New York: IEEE.
33.
ParkDKimHKempCC. Multimodal anomaly detection for assistive robots. Auton Robots2019; 43: 611–629.
34.
BeggelLPfeifferMBischlB. Robust anomaly detection in images using adversarial autoencoders. In: BrefeldUFromontEHothoA, et al. (eds.) Joint European Conference on machine learning and knowledge discovery in databases. Cham: Springer, 2019, pp. 206–222.
35.
FlynnEBToddMDWilcoxPD, et al. Maximum-likelihood estimation of damage location in guided-wave structural health monitoring. P Roy Soc A: Math Phys Eng Sci2011; 467: 2575–2596.
36.
SmithGDe FreitasJFRobinsonT, et al. Speech modelling using subspace and EM techniques. In: Proceedings of the advances in neural information processing systems, Denver, CO, 29 Noevember–4 December 2000, pp. 796–802.
37.
WieselAEldarYCBeckA. Maximum likelihood estimation in linear models with a Gaussian model matrix. IEEE Signal Process Lett2006; 13: 292–295.
38.
MaJWuOHuangB, et al. Expectation maximization estimation for a class of input nonlinear state space systems by using the Kalman smoother. Signal Process2018; 145: 295–303.
39.
GhahramaniZHintonGE. Parameter estimation for linear dynamical systems. Technical Report, Report No. CRG-TR-96-2, 22 February 1996. Toronto, ON: Department of Computer Science, University of Toronto.
40.
GuoFWuODingY, et al. A data-based augmented model identification method for linear errors-in-variables systems based on EM algorithm. IEEE Trans Ind Electron2017; 64: 8657–8665.
41.
BarberD. Bayesian reasoning and machine learning. Cambridge: Cambridge University Press, 2012.
JinSZhangZChakrabartyK, et al. Changepoint-based anomaly detection for prognostic diagnosis in a core router system. IEEE Trans Comput-aid Design Integ Circu Syst2018; 38: 1331–1344.
44.
BaghdasaryanAGBeexA. Automatic phoneme recognition with segmental hidden Markov models. In: Proceedings of the 2011 conference record of the forty fifth Asilomar conference on signals, systems and computers (ASILOMAR), Pacific Grove, CA, 6–11 November 2011, pp. 569–574. New York: IEEE.
45.
Pruteanu-MaliniciICarinL. Infinite hidden Markov models for unusual-event detection in video. IEEE Trans Image Process2008; 17: 811–822.
46.
JeongJParkEHanWS, et al. Identifying outliers of non-Gaussian groundwater state data based on ensemble estimation for long-term trends. J Hydrol2017; 548: 135–144.
47.
HuangZChalabiZ. Use of time-series analysis to model and forecast wind speed. J Wind Eng Ind Aerodyn1995; 56: 311–322.
48.
WanH-PNiY-Q. Bayesian modeling approach for forecast of structural stress response using structural health monitoring data. J Struct Eng2018; 144: 04018130.
YiT-HHuangH-BLiH-N. Development of sensor validation methodologies for structural health monitoring: a comprehensive review. Measurement2017; 109: 200–214.
51.
GuoY. Monitoring-based assessment of bridges subject to ship collision. Hung Hom: The Hong Kong Polytechnic University, 2010.
52.
WanH-PNiY-Q. Bayesian multi-task learning methodology for reconstruction of structural health monitoring data. Struct Health Monit2019; 18: 1282–1309.
