Search

Track geometry is one of the critical indicators of railway tracks' condition which requires continuous monitoring and maintenance over time. In this paper, a novel artificial intelligence (AI) based framework is proposed for railway track geometry inspection using vibration data collected from a dedicated measuring high-speed train. This AI-based anomaly track detection approach consists of two main stages. Firstly, a subset of features that best characterizes the track condition is defined. Several dynamic features from time domain data are extracted and importance scores are assigned to them, to determine the most effective subset for the purpose of track condition monitoring. Secondly, a data-driven based anomaly detection approach is developed to assess and identify track geometrical defects. In this stage, the acceleration responses collected from an in-service train traversing on a healthy track zone are employed as input into a One-Class Support Vector Machine (OCSVM) algorithm. The proposed algorithm defines the anomalies as relative changes to the historical behaviour. A comprehensive dataset from field measurements using a Société Nationale des Chemins de Fer Français (SNCF) Réseau IRIS320 highspeed train is used in this paper to implement the proposed approach. In addition, the impact of using different features and different locations/directions of the sensors on the accuracy of detecting geometrical defects is investigated. It is also shown that the OCSVM approach outperforms other algorithms based on Isolation Forest (IF), Local Outlier Factor (LOF), and Robust Mahalanobis Distance (RMD) in terms of recall, precision, and F1-score. The proposed anomaly detection approach has demonstrated a 12% increase in defect detection accuracy compared to the direct utilization of the raw acceleration response, which can facilitate track monitoring using in-service trains while providing cost-efficient maintenance in the future. [Display omitted] [ABSTRACT FROM AUTHOR]

Track geometry is one of the critical indicators of railway tracks' condition which requires continuous monitoring and maintenance over time. In this paper, a novel artificial intelligence (AI) based framework is proposed for railway track geometry inspection using vibration data collected from a dedicated measuring high-speed train. This AI-based anomaly track detection approach consists of two main stages. Firstly, a subset of features that best characterizes the track condition is defined. Several dynamic features from time domain data are extracted and importance scores are assigned to them, to determine the most effective subset for the purpose of track condition monitoring. Secondly, a data-driven based anomaly detection approach is developed to assess and identify track geometrical defects. In this stage, the acceleration responses collected from an in-service train traversing on a healthy track zone are employed as input into a One-Class Support Vector Machine (OCSVM) algorithm. The proposed algorithm defines the anomalies as relative changes to the historical behaviour. A comprehensive dataset from field measurements using a Société Nationale des Chemins de Fer Français (SNCF) Réseau IRIS320 highspeed train is used in this paper to implement the proposed approach. In addition, the impact of using different features and different locations/directions of the sensors on the accuracy of detecting geometrical defects is investigated. It is also shown that the OCSVM approach outperforms other algorithms based on Isolation Forest (IF), Local Outlier Factor (LOF), and Robust Mahalanobis Distance (RMD) in terms of recall, precision, and F1-score. The proposed anomaly detection approach has demonstrated a 12% increase in defect detection accuracy compared to the direct utilization of the raw acceleration response, which can facilitate track monitoring using in-service trains while providing cost-efficient maintenance in the future. [Display omitted] [ABSTRACT FROM AUTHOR]

This paper proposes a novel data-driven framework for scour detection around offshore wind turbines (OWTs), where damage features are derived from wind and wave-induced acceleration signals collected along the tower. A numerical model of the NREL 5 MW wind turbine, which considers aerodynamic and hydrodynamic loading with soil-structure interaction (SSI) and servo-dynamics, is developed. The model is used to simulate the acceleration responses along the tower for a healthy structure, and a structure affected by progressive scour. A data segmentation process is initially performed on the collected data, which is followed by a feature selection scheme based on the analysis-of-variance (ANOVA) algorithm, to eliminate irrelevant characteristics from the time domain feature set of responses. The proposed framework consists of two main components: (a) offline training, and (b) real-time classification. The acceleration responses collected from the healthy structure and the structure subjected to three different damage scenarios (different scour depths) and under various load conditions, are used in the offline training mode. The selected feature vector from the feature extraction process is used as input to a Naive Bayes classifier (NBC) algorithm to train the model. In the real-time classification, a prediction of the scour depth affecting the structure is performed using a new dataset simulated from unseen load cases and scour conditions of the OWT. The results show that the model trained in the offline stage can predict the scour depth in the real-time monitoring stage with performance measures over approximately 94%. • A novel data-driven framework is proposed for scour detection around offshore wind turbines. • Damage features are derived from wind and wave-induced acceleration signals collected along the tower. • The data from the healthy structure and the structure subjected to three different damage scenarios and under various load conditions, are used in the offline training mode. • A prediction of the scour depth is performed using a new dataset simulated from unseen load cases and scour conditions of the OWT in the real-time classification. [ABSTRACT FROM AUTHOR]

This paper proposes a novel optimization framework for the optimal design of Tuned Mass Damper Inerter (TMDI) devices to mitigate the excess vibration of multi degree of freedom (MDOF) buildings under earthquake ground motions. Minimizing the H ∞ - norm of the top floor displacement transfer functions is considered as the objective function for the optimization process. A Tuned Mass Damper Inerter (TMDI) is employed in which a TMD is placed at the top floor of a building connected via inerter. The top floor displacement of the building under eighteen (18) real earthquake ground motions classified in terms of low, intermediate and high frequency contents, is minimized using three optimization algorithms, particle swarm optimization (PSO), slime mould algorithm (SMA) and Harris Hawks optimization (HHO). The effectiveness of different algorithms is assessed using the RMS (root mean square) based values and the reduction ratio. The optimised parameters of the TMDI (mass ratio, frequency ratio, damping ratio and inertance ratio) obtained using the slime mould algorithm (SMA) are found to be better than the other optimization algorithms for the numerical example carried out in this paper. The results show that, the optimised TMDI has significant effect on the suppression of top floor displacement, acceleration with a lightweight TMD. The optimal parameters of TMDI shows a remarkable potential in suppressing maximum displacement reduction ratios of 0.83, 0.85 and 0.86 in response to low, intermediate and high frequency content earthquake ground motion, respectively. Based on the numerical results, the passive TMDI unveils a promising vibration calming system for damped elastic structure. • H ∞ norm transfer function is used to evaluate TMDI parameters. • Three novel optimization algorithms such as PSo, SMA and HHO are compared to find damper parameters. • The effectiveness of optimised TMDI under different frequency content earthquake is evaluated. [ABSTRACT FROM AUTHOR]

Highlights • An effective strategy to use metamodels (surrogate models), for model updating in structural health monitoring. • A comparative study of ten most common metamodeling techniques used for damage detection and the severity of damage; the first such study in SHM research. • A novel optimization algorithm, Colliding Bodies, CBO, with and without surrogate models for damage severity and computational efficiency. • Proposed application of this approach for general hysteretic degradation of a structure subjected to external excitations. • Identified LS-SVM as the most promising tool, combined with metamodels, for damage severity assessment. Abstract Despite advances in computer capacity, the enormous computational cost of running complex engineering simulations makes it impractical to rely exclusively on simulation for the purpose of structural health monitoring. To cut down the cost, surrogate models, also known as metamodels, are constructed and then used in place of the actual simulation models. In this study, structural damage detection is performed using two approaches. In both cases ten popular metamodeling techniques including Back-Propagation Neural Networks (BPNN), Least Square Support Vector Machines (LS-SVMs), Adaptive Neural-Fuzzy Inference System (ANFIS), Radial Basis Function Neural network (RBFN), Large Margin Nearest Neighbors (LMNN), Extreme Learning Machine (ELM), Gaussian Process (GP), Multivariate Adaptive Regression Spline (MARS), Random Forests and Kriging are used and the comparative results are presented. In the first approach, by considering dynamic behavior of a structure as input variables, ten metamodels are constructed, trained and tested to detect the location and severity of damage in civil structures. The variation of running time, mean square error (MSE), number of training and testing data, and other indices for measuring the accuracy in the prediction are defined and calculated in order to inspect advantages as well as the shortcomings of each algorithm. The results indicate that Kriging and LS-SVM models have better performance in predicting the location/severity of damage compared with other methods. In the second approach, after locating precisely the eventual damage of a structure using modal strain energy based index (MSEBI), to efficiently reduce the computational cost of model updating during the optimization process of damage severity detection, the MSEBI of structural elements is evaluated using a properly trained surrogate model. The results indicate that after determining the damage location, the proposed solution method for damage severity detection leads to significant reduction of computational time compared to finite element method. Furthermore, engaging colliding bodies optimization algorithm (CBO) by efficient surrogate model of finite element (FE) model, maintains the acceptable accuracy of damage severity detection. [ABSTRACT FROM AUTHOR]

• Propose a novel deep learning model based on Runge-Kutta recurrent neural network (RKRNN) with prior knowledge to realize structural system identification and seismic response prediction. • Formulate a partition training strategy to train the proposed neural network to improve the efficiency of training. • Utilize three numerical examples to valid the feasibility of RKRNN model including a linear three degrees of freedom system, a nonlinear single degree of freedom system with Bouc-Wen hysteresis and a simply supported bridge. • Site monitoring data from a bridge located in California has been used to further validate the proposed approach. In the seismic analysis of structural systems, dynamic response prediction is an essential problem and is significant in every stage during the structural life cycle. Conventionally, response analysis is carried out by numerical analysis. However, when the structural parameter is unknown, the establishment of a numerical model will be difficult. Enlightened by the Runge-Kutta (RK) numerical algorithm, this paper proposes a novel recurrent neural network named Runge-Kutta recurrent neural network (RKRNN) to realize the seismic response prediction. A partition training strategy is formulated to train the proposed neural network and to improve the efficiency of training. The proposed model can be trained by using a limited number of samples. Three numerical examples are utilized to validate the feasibility of RKRNN model including a linear three degrees of freedom (DOFs) system, a nonlinear single DOF system with Bouc-Wen hysteresis, and a numerical reinforced concrete bridge model. Additionally, the site monitoring data from a real-world bridge is utilized to further validate the proposed network. The results show that the proposed RKRNN model can effectively and efficiently predict the structural response under seismic load and exhibits robustness to noise, with good potential for applications in engineering practice. [ABSTRACT FROM AUTHOR]