Data-Driven Railway Crosstie Support Condition Prediction Using Deep Residual Neural Network: Algorithm and Application

Ballasted track substructure is designed and constructed to provide uniform crosstie support and serve the functions of drainage and load distribution over trackbed. Poor and nonuniform support conditions can cause excessive crosstie vibration which will negatively affect the crosstie flexural bending behavior. Furthermore, ballast–tie gaps and large contact forces at the crosstie–ballast interface will result in accelerated ballast layer degradation and settlement accumulation. Inspection of crosstie support condition is therefore necessary while very challenging to implement using current methods and technologies. Based on deep learning artificial intelligence techniques and a developed residual neural network (ResNet), this paper introduces an innovative data-driven prediction approach for crosstie support conditions as demonstrated from a full-scale ballasted track laboratory experiment. The discrete element method (DEM) is leveraged to provide training and testing data sets for the proposed prediction model. K-means clustering is applied to establish ballast layer subsections with representative ballast particles and provide additional insights on layer zoning for dynamic behavior trends. When provided with DEM simulated particle vertical accelerations, the proposed deep learning ResNet could achieve 100% training and 95.8% testing accuracy. Fed with vertical acceleration measurements captured by advanced “SmartRock” sensors from a full-scale ballasted track laboratory experiment, the trained model could successfully reach a high accuracy of 92.0%. Based on the developed deep learning approach and the research findings presented in this paper, the innovative crosstie support condition prediction system is envisioned to provide railroaders accurate, timely, and repeatable inspection and monitoring opportunities without disrupting railway network operations.

Determination of Mapping Relation between Wheel Polygonalisation and Wheel/Rail Contact Force for Railway Freight Wagon Using Dynamic Simulation

The polygonal wear around the wheel circumference could pose highly adverse influences on the wheel/rail interactions and thereby the performance of the vehicle system. In this study, the effects of wheel polygonalisation on the dynamic responses of a freight wagon are investigated through development and simulations of a comprehensive coupled vehicle-track dynamic model. The model integrates flexible ballasted track and wheelsets subsystem models so as to account for elastic deformations caused by impact loads induced by the wheel polygonalisation. Subsequently, the vehicles with low-order polygonal wear, whether in empty or loaded conditions, are simulated at different speeds considering different amplitudes and harmonic orders of the wheel polygonalisation and thus the mapping relation between wheel/rail impact force and wheel polygonalisation is obtained. The results reveal that the low-order wheel polygonalisation except 1st order and 3rd order can give rise to high-frequency impact loads at the wheel/rail interface and excite 1st-bend modes of the wheelset and “P2 resonance” leading to high-magnitude wheel/rail contact force at the corresponding speed.

Three-Dimensional Vehicle–Curved Track Dynamic Model Based on FEM and DEM

In this work, a systematic vehicle–curved track dynamic model is presented, in which the vehicle is modeled as a multi-rigid-body system. The track structure is modeled by finite element method with curved rail beam element considered in geometry. To obtain accurate dynamic behavior of railway ballasted track, the resistance characteristics of ballast bed are revealed by introducing the discrete element method. Besides, a two-step iterative-update method is improved to solve the multi-nonlinearity of the vehicle–curved track dynamic interaction. To improve the computational efficiency, and the improved infinite cyclic calculation method is introduced. Apart from the model validations, the application of this model in engineering practices, such as the vehicle-induced vibration of the continuously welded rail (CWR), has been revealed, and some conclusions are drawn from the numerical studies.

Numerical Analysis of Additional Stresses in Railway Track Elements Due to Subgrade Settlement Using FEM Simulation

The prediction of additional stresses in ballasted track due to subgrade deformation is the main objective of the present paper. In this context, a 2D finite element model of ballasted railway track was built using the ANSYS Workbench program. Based on this model, an investigation of stresses and deformation values of track elements was conducted in three cases with different contact types. It was found that the case introducing the status of a new track, which has frictional contacts between sleepers and ballast with bonded contacts between other elements, has lower stresses in most of the track elements. Moreover, this case was applied for studying the effect of the settlement on track elements. It was found that stresses increased with increasing the settlement value. The average percentages of increased stresses are 4.18%, 5.85%, and 7.21% in railhead, tie plate, and sleeper, respectively, due to a 1 mm increase in the settlement. Finally, a second-degree polynomial equation was derived to predict the additional stresses in each element due to track settlement. It is expected that this study would help to decrease the maintenance costs and extend the service life of the track elements by predicting the additional stresses in them.