An ensemble-based change-point detection method for identifying unexpected behaviour of railway tunnel infrastructures

Abstract

A large amount of data is generated by Structural Health Monitoring (SHM) systems and, as a consequence, processing and interpreting this data can be difficult and time consuming. Particularly, if work activities such as maintenance or modernization are carried out on a bridge or tunnel infrastructure, a robust data analysis is needed, in order to accurately and quickly process the data and provide reliable information to decision makers. In this way the service disruption can be minimized and the safety of the asset and the workforce guaranteed. In this paper a data mining method for detecting critical behaviour of a railway tunnel is presented. The method starts with a pre-processing step that aims to remove the noise in the recorded data. A feature definition and selection step is then performed to identify the most critical area of the tunnel. An ensemble of change-point detection algorithms is proposed, in order to analyse the critical area of the tunnel and point out the time when unexpected behaviour occurs, as well as its duration and location. The work activities, which are carried out at the time of occurrence of the critical behaviour and have caused this behaviour, are finally identified from a database of the work schedule and used for the validation of the results. Using the proposed method, fast and reliable information about infrastructure condition is provided to decision makers.

An automatic bridge damage diagnostics method using empirical mode decomposition based health indicators and neuro‐fuzzy classification

“…The SHM of bridges is of particular interest because in Europe there are more than one million of bridges, and this number is expected to increase 5 . Bridges need SHM strategies to (i) identify the ongoing degradation mechanisms of the bridge materials and avoid unexpected and catastrophic failure of the asset 6 and to (ii) ensure the safety of the workforce and understand whether the infrastructure behaviour during the work activities (such as maintenance and renewal activities) is within the predicted safety limits 7 . The behaviour of a bridge is monitored by installing a measurement system (such as GPS receivers, accelerometers, strain gauges and cameras) on the bridge infrastructure 8–10 .…”

“…5 Bridges need SHM strategies to (i) identify the ongoing degradation mechanisms of the bridge materials and avoid unexpected and catastrophic failure of the asset 6 and to (ii) ensure the safety of the workforce and understand whether the infrastructure behaviour during the work activities (such as maintenance and renewal activities) is within the predicted safety limits. 7 The behaviour of a bridge is monitored by installing a measurement system (such as GPS receivers, accelerometers, strain gauges and cameras) on the bridge infrastructure. [8][9][10] A large amount of data is generated by these sensors continuously, and thus, data-driven methods are required to assess the health state of the infrastructure automatically, accurately and rapidly.…”

A change-point detection method for detecting and locating the abrupt changes in distributions of damage-sensitive features of SHM data, with application to structural condition assessment

“…Change-point detection originates from statistical quality control, now it has become an important research topic in statistics 8 and is increasingly recognized in the engineering field as a highly effective tool for damage detection, fault detection, abnormal structural behavior detection, and so on. [9][10][11][12][13][14][15][16][17] In change-point detection, the decision on a change event is generally made according to the hypothesis testing under a given significance level; thus, the change-point detection technique has more potential in revealing subtler changes that might correspond to minor damage at an early stage. So far, various change-point detection methods have been developed or adopted for different SHM applications.…”

“…Most recently, Li et al 15 developed a change-point detection-based method for detecting the abnormal changes in crack behavior of concrete dams. Additionally, change-point detection methods have also been successfully applied in tunnel monitoring for abnormal structural behavior identification 16 as well as in high-speed rail monitoring for large deformation detection. 17 To date, the change-point detection methods employed for SHM are generally applicable to scalar or vector data.…”