Rail Corrugation Detection and Characterization Using Computer Vision

Rail corrugation appears as oscillatory wear on the rail surface caused by the interaction between the train wheels and the railway. Corrugation shortens railway service life and forces early rail replacement. Consequently, service can be suspended for days during rail replacement, adversely affecting an important means of transportation. We propose an inspection method for rail corrugation using computer vision through an algorithm based on feature descriptors to automatically distinguish corrugated from normal surfaces. We extract seven features and concatenate them to form a feature vector obtained from a railway image. The feature vector is then used to build support vector machine. Data were collected from seven different tracks as video streams acquired at 30 fps. The trained support vector machine was used to predict test frames of rails as being either corrugated or normal. The proposed method achieved a high performance, with 97.11% accuracy, 95.52% precision, and 97.97% recall. Experimental results show that our method is more effective in identifying corrugated images than reference state-of the art works.

Parametric Study of Track Support Structure to Minimize the Occurrence Possibility of Rail Corrugation

Changing the track support structure is an effective method to suppress or eliminate rail corrugation in practical engineering. Rail corrugation on small-radius curves with booted short sleepers is the main research object in the present paper. A relevant finite element model of the wheelset-track system supported by booted short sleepers is built combined with the dynamic analysis of the vehicle-track system. The effects of various parameters of booted short sleeper structure on the wheel–rail friction-induced vibration are investigated by complex eigenvalue analysis. Considering the interaction of multiple parameters in the booted short sleeper structure, the multi-parameter fitting equation forecasting the possibility of rail corrugation is obtained using the least squares algorithm. Results show that wheel–rail friction-induced oscillation is a contributing factor in the formation of rail corrugation. Controlling wheel–rail friction-induced oscillation with a frequency of about 300 Hz is beneficial to suppress the possibility of rail corrugation in sections with booted short sleepers. Lower fastener stiffness or greater vertical fastener damping make it less likely that rail corrugation will occur. Rail corrugation is not generated when the vertical stiffness of the fastener is controlled below 20 MN/m in the booted short sleeper.

Field Measurement Analysis and Control Measures Evaluation of Metro Vehicle Noise Caused by Rail Corrugation

The noise caused by rail corrugation seriously affects the operation quality of metro vehicles. In this work, the rail corrugation, interior noise and wheel–rail noise of a metro line were tested, and the test results were compared with those after two kinds of treatments. The results show that rail corrugation is the main cause of the abnormal interior noise. The interior noise in many sections exceeds the limit, where clear rail corrugations from 31.5~63 mm are found. When the train passes through the rail corrugation section, the interior noise shows a clear increase, and the maximum increase is higher than 25 dB(A). After increasing the lateral stiffness of the track and rail grinding, the interior noise is reduced by 11.4 dB(A). After a long renovation time, the interior noise is effectively equal to that when the renovation was completed. The research results of this work can provide a reference for rail corrugation treatment and noise control.

Spectral Characteristics of Rail Surface by Measuring the Growth of Rail Corrugation

Continuous interaction between wheels and rails during train operation results in rail wear and tear. Corrugation of the rail surface is particularly caused by the contact mechanism between train wheel and rail and increases the vibration and dynamic wheel load, and if continued, leads to various defects and breakage of the track. Many devices are used to measure corrugation accurately, but measurement deviation varies greatly by measuring device. The most common corrugation measurement system measures surface roughness with a vibration acceleration sensor or displacement sensor. Corrugation with different pitches can be calculated by assuming the longitudinal rail surface as a chord with variable wavelength. Recent systems use a measurement model applying multiple sensors for more accurate measurement. This study investigated spectral characteristics of rail surface roughness based on long time measurement results. In particular, spectral changes upon the growth of rail surface wear and corrugation were analyzed when using the measurement system applying the chord offset method with multiple sensors. The results of analysis were verified through the field subway section, a running track undergoing corrugation, three months after initial measurement, and compared the measurement results according to the number of sensors. Additionally, the rail surface wavelength bands that affected measurement value according to the number of sensors were analyzed.