Advances in Condition Monitoring of Railway Infrastructure.

In recent years, there has been a notable surge in investments directed towards developing new railway lines and revitalising existing ones, reflecting a global commitment to enhance transportation infrastructure [...].

A Machine-Learning-Based Approach for Railway Track Monitoring Using Acceleration Measured on an In-Service Train.

In this paper, a novel railway track monitoring approach is proposed that employs acceleration responses measured on an in-service train to detect the loss of stiffness in the track sub-layers. An Artificial Neural Network (ANN) algorithm is developed that works with the energies of the train acceleration responses. A numerical model of a half-car train coupled with a track profile is employed to simulate the train vertical acceleration. The energy of acceleration signals measured from 100 traversing trains is used to train the ANN for healthy track conditions. The energy is calculated every 15 m along the track, each of which is called a slice. In the monitoring phase, the trained ANN is used to predict the energies of a set of train crossings. The predicted energies are compared with the simulated ones and represented as the prediction error. The damage is modeled by reducing the soil stiffness at the sub-ballast layer that represents hanging sleepers. A damage indicator (DI) based on the prediction error is proposed to visualize the differences in the predicted energies for different damage cases. In addition, a sensitivity analysis is performed where the impact of signal noise, slice sizes, and the presence of multiple damaged locations on the performance of the DI is assessed.

Indirect Monitoring of Frequencies of a Multiple Span Bridge Using Data Collected from an Instrumented Train: A Field Case Study.

In this paper, a field study is carried out to monitor the natural frequencies of Malahide viaduct bridge which is located in the north of Dublin. The bridge includes a series of simply supported spans, two of which collapsed in 2009 and were replaced. The replaced spans are stiffer than most of the others and these differences resulted in higher natural frequencies. An indirect bridge monitoring approach is employed in which acceleration responses from an instrumented train are used to estimate the natural frequencies of each span of the viaduct showing the locations of the two replaced spans with higher stiffness. For the indirect approach, an Ensemble Empirical Mode Decomposition (EEMD)-based Hilbert Huang Transform (HHT) technique is employed to identify the natural frequency of each span. This is carried out by analysing the Instantaneous Frequencies (IFs) from the calculated intrinsic mode functions. The average of the IFs calculated using 41 runs of the instrumented train (with varying carriage mass and speed for each run) are used to estimate the natural frequencies. To assess the feasibility of the indirect approach, a bespoke set of direct measurements was taken using accelerometers attached successively on each span of the viaduct. The free and forced vibrations from each span are used to estimate the first natural frequencies. The frequencies obtained from drive-by measurements are compared to those from direct measurements which confirms the effectiveness of indirect approaches. In addition, the instantaneous amplitudes of the drive-by signals are used to indicate the location of the stiffer spans. Finally, the accuracy and robustness of the indirect approaches for monitoring of multi span bridges are discussed.

A Machine Learning Approach to Bridge-Damage Detection Using Responses Measured on a Passing Vehicle.

This paper proposes a new two-stage machine learning approach for bridge damage detection using the responses measured on a passing vehicle. In the first stage, an artificial neural network (ANN) is trained using the vehicle responses measured from multiple passes (training data set) over a healthy bridge. The vehicle acceleration or Discrete Fourier Transform (DFT) spectrum of the acceleration is used. The vehicle response is predicted from its speed for multiple passes (monitoring data set) over the bridge. Root-mean-square error is used to calculate the prediction error, which indicates the differences between the predicted and measured responses for each passage. In the second stage of the proposed method, a damage indicator is defined using a Gaussian process that detects the changes in the distribution of the prediction errors. It is suggested that if the bridge condition is healthy, the distribution of the prediction errors will remain low. A recognizable change in the distribution might indicate a damage in the bridge. The performance of the proposed approach was evaluated using numerical case studies of vehicle-bridge interaction. It was demonstrated that the approach could successfully detect the damage in the presence of road roughness profile and measurement noise, even for low damage levels.

Scour Damage Detection and Structural Health Monitoring of a Laboratory-Scaled Bridge Using a Vibration Energy Harvesting Device.

A vibration-based bridge scour detection procedure using a cantilever-based piezoelectric energy harvesting device (EHD) is proposed here. This has an advantage over an accelerometer-based method in that potentially, the requirement for a power source can be negated with the only power requirement being the storage and/or transmission of the data. Ideally, this source of power could be fulfilled by the EHD itself, although much research is currently being done to explore this. The open-circuit EHD voltage is used here to detect bridge frequency shifts arising due to scour. Using one EHD attached to the central bridge pier, both scour at the pier of installation and scour at another bridge pier can be detected from the EHD voltage generated during the bridge free-vibration stage, while the harvester is attached to a healthy pier. The method would work best with an initial modal analysis of the bridge structure in order to identify frequencies that may be sensitive to scour. Frequency components corresponding to harmonic loading and electrical interference arising from experiments are removed using the filter bank property of singular spectrum analysis (SSA). These frequencies can then be monitored by using harvested voltage from the energy harvesting device and successfully utilised towards structural health monitoring of a model bridge affected by scour.