Anomaly detection for large span bridges during operational phase using structural health monitoring data

Abstract

In view of the limitation of damage detection in practical applications for large scale civil structures, a practical method for anomaly detection is developed. Within the anomaly detection framework, wavelet transform and generalized Pareto distribution are adopted for data processing. In detail, to reduce the influence of thermal responses on signal fluctuations induced by anomaly events, wavelet transform is employed to separate thermal effects from raw signals based on the distinguished frequency bandwidths. Subsequently, a two-level anomaly detection method is proposed, i.e., threshold-based anomaly detection and anomaly trend detection. For the threshold-based anomaly detection, the threshold for anomaly detection is determined by generalized Pareto distribution analytics, corresponding to a 95% guarantee rate within 100 years. Moreover, the threshold is periodically updated by incorporating the latest monitoring data to model the increase of traffic volumes and gradual degradations of structures. For the anomaly trend detection, the moving fast Fourier transform is adopted for discussion. Finally, the mid-span deflection of Xihoumen Suspension Bridge is selected as the index to validate the effectiveness of the proposed methodology. Two types of anomaly events are assumed in the case study, i.e., the overloading event and structural damage. The two-level anomaly detection is implemented. It is indicated through the case study that the proposed anomaly detection approach (without the influence of temperature) is able to detect three 100-ton overloaded vehicles and damages in main cables. However, the assumed cases subject to 100-ton vehicle and damages in stiffening girders are hardly detected by using the deflection index, owing to the sensitivity of the index to each anomaly event. In the future studies, a structural health monitoring-based multi-index anomaly detection system is promising to ensure the operational and structural safety of large span bridges.

Probabilistic anomaly trend detection for cable-supported bridges using confidence interval estimation

“…In existing static response-based anomaly detection investigations, physical quantities (e.g., deflection) and their changes in form (e.g., cointegration residual) were always adopted as anomaly detection indexes owing to their straightforwardness and practicability (Tome et al, 2020; Xu et al, 2020b). However, these indexes are sensitive to signal spikes which are common phenomena for measurements of SHM systems, leading to false detection.…”

“…Similarly, the energy indexes are calculated and plotted in Figure 21 together with triggers of point estimation and the four confidence levels. The details regarding calculation of the trigger using point estimation method is presented in our previous paper (Xu et al, 2020b). The trigger derived from the point estimation is lower than that of 20% confidence level, which is prone to raise false detection.…”

“…Considering the influence of thermal effects to static parameter-based anomaly detection, various temperature-driven anomaly detection approaches are developed. Xu et al (2020b) proposed a two-level anomaly detection for a suspension bridge using girder deflection measurements, where multi-resolution wavelet-based method was used to separate thermal responses from recorded signals. Zhu et al (2019) introduced the blind source separation to improve the performance of moving principal component analysis for anomaly detection by using strain data, where three cases were used to verify the effectiveness of the proposed methodology.…”

Estimation of Extreme Cable Forces of Cable‐Stayed Bridges Based on Monitoring Data and Random Vehicle Models

“…For instance, China's annual growth rate of freight traffic volume approached 6% in 2018 [5], which was a critical parameter for evaluation of structural performances. Structural responses might exceed the designed ones in the following several decades owing to the pressure of rapid increase in traffic volumes [6]. is phenomenon will inevitably impact operational and structural safety of long-span cable-stayed bridges.…”

Functional Kernel Density Estimation: Point and Fourier Approaches to Time Series Anomaly Detection

“…Being able to detect anomalies has many applications, including in the fields of medicine and healthcare management [ 1 , 2 ]; in data acquisition, such as filtering out anomalous readings [ 3 ]; in computer security [ 4 ]; in media monitoring [ 5 ]; and many in the realm of public safety such as identifying thermal anomalies that may precede earthquakes [ 6 ], identifying potential safety issues in bridges over time [ 7 ], detecting anomalous conditions for trains [ 8 ], system level anomaly detection among different air fleets [ 9 ], and identifying which conditions pose increased risk in aviation [ 10 ]. Given a dataset, anomaly detection is about identifying individual data that are quantitatively different from the majority of other members of the dataset.…”