No abstract

In this paper, we propose a deep convolutional neural network solution to the analysis of image data for the detection of rail surface defects. The images are obtained from many hours of automated video recordings. This huge amount of data makes it impossible to manually inspect the images and detect rail surface defects. Therefore, automated detection of rail defects can help to save time and costs, and to ensure rail transportation safety. However, one major challenge is that the extraction of suitable features for detection of rail surface defects is a non-trivial and difficult task. Therefore, we propose to use convolutional neural networks as a viable technique for feature learning. Deep convolutional neural networks have recently been applied to a number of similar domains with success. We compare the results of different network architectures characterized by different sizes and activation functions. In this way, we explore the efficiency of the proposed deep convolutional neural network for detection and classification. The experimental results are promising and demonstrate the capability of the proposed approach.Accepted Author Manuscript. Link to published article (IEEE): http://dx.

No abstract

Squats are a type of short-wave rolling contact fatigue defect whose early detection can contribute to cost reductions in the railway industry. This paper demonstrates how the early detection of squats is possible via enhanced instrumentation based on axle box acceleration (ABA) and adequate postprocessing. Three improvements are discussed. The first corresponds to the use of longitudinal ABA to enhance measurement sensitivity to light squats. Compared to vertical ABA, longitudinal ABA does not contain the vibrations of the rail, fastening, sleepers, and ballast, and thus, the impact-related vibration is considerably stronger in the signal. The second improvement considers the use of multiple sensors, noise-reduction techniques, and repeated measurements. Due to hunting, the wheels of a measuring train do not always pass over small squats; thus, light squats are more likely to be detected using multiple sensors and multiple measurement runs. The third improvement concerns the signal-processing solution for the reduction of disturbances from wheel defects. Extensive field measurements show that these improvements make the characteristics of squats more visible in signals and allow the squats to be distinguished from vibrations of other origins.

Abstract-This paper reviews different observer design methods for linear dynamic distributed-parameter systems. In such systems, the states, inputs, and outputs depend on some spatial variable. This dependence, along with additional aspects such as the boundary conditions, increase the complexity of the state estimation problem and of the design methods. The paper in particular surveys observers for first-order and second-order linear distributed-parameter systems based on their infinitedimensional and finite-dimensional descriptions.