No abstract

No abstract

The railroad ballast layer consists of discrete aggregate particles, and the discrete element method (DEM) is the most widely adopted numerical method to simulate the particulate nature of ballast materials and their particle interactions. Large-scale triaxial tests performed in the laboratory under controlled monotonic and repeated loading conditions are commonly considered the best means to measure macroscopic mechanical properties of ballast materials, such as strength, modulus, and deformation characteristics, directly related to load-carrying and drainage functions of the ballast layer in the field. A DEM modeling approach is described for railroad ballast with realistic particle shapes developed from image analysis to simulate large-scale triaxial compression tests on a limestone ballast material. The ballast DEM model captures the strength behavior from both the traditional slow and the rapid shear loading rate types of monotonic triaxial compression tests. The results of the experimental study indicated that the shearing rate had insignificant influence on the results of the triaxial compression tests. The results also showed that the incremental displacement approach captured the measured shearing response, yet could save significant computational resources and time. This study shows that the DEM simulation approach combined with image analysis has the potential to be a quantitative tool to predict ballast performance.

No abstract

Rail surface defects (RSDs) are a major problem that reduces operation safety. Unfortunately, the existing RSD detection systems have very limited accuracy. Current image processing methods are not tailored for the railway track and many fully convolutional networks (FCN)-based methods suffer from the blurry rail edges (RE). This paper proposes a new rail boundary guidance network (RBGNet) for salient RS detection. First, a novel architecture is proposed to fully utilize the complementarity between the RS and the RE to accurately identify the RS with well-defined boundaries. The newly developed RBGNet injects high-level RS object information into shallow RS edge features by a progressive fused way for obtaining fine edge features. Then, the system integrates the refined edge features with RS features at different high-level layers to predict the RS precisely. Second, an innovative hybrid loss consisting of binary cross entropy (BCE), structural similarity index measure (SSIM), and intersection-over-union (IoU) is proposed and equipped into the RBGNet to supervise the network and learn the transformation between the input and ground truth. The input and ground truth then further refine the RS location and edges. Conveniently, an image-based model for RSD detection and quantification is also developed and integrated for an automatic inspection purpose. Finally, experiments conducted on the complex unmanned aerial vehicle (UAV) rail dataset indicate the system can achieve a high detection rate with good adaptation capability in complicated environments. INTRODUCTIONThe rapid growth of the railroad network has put tremendous pressure on track inspection and maintenance. As of 2020, United States has over 250,000 km of railroad track, which is the biggest network in the world (Railway Technology, 2020). China operates about 141,400 km of track, ranking the second in the world, while its 36,000 km of high-speed track is the most comprehensive high-speed © 2021 Computer-Aided Civil and Infrastructure Engineering passenger service network in the world (Xinhuanet, 2020). Russia and India rank third and fourth in terms of the track mileage with over 85,500 km and 65,000 km of track, respectively. Rail breakage, rail defects, and derailment are the leading factors of train accidents (Guo et al., 2021;Sharma et al., 2018). Specifically, it is reported that around 90% of railway derailment accidents can be related to rail defects (AlNaimi, 2020). In general, rail surface defects (RSDs) reference to the loss of materials on the rail head