Development of the rail corrugation detection technology using an image analysis technique

초록

Rail corrugation, a form of wave wear that occurs at the head of the rail due to various causes, not only shortens the rail life, but also is known as a major cause of abnormal noise from running trains and low passenger comfort. The damage by a change in the wheel-rail contact force is known as one of the mechanisms generating the rail corrugation, but there is a limitation in predicting the actual cause of the corrugation due to a variety of influence factors. Therefore, the occurrence of rail corrugation can be detected when it is developed enough to be seen with the naked eye during track inspection, when abnormal noise and low passenger comfort occur, or when operating the regular track measurement vehicle. The detection techniques include using a sensor to measure displacement and the acceleration by contacting the rail surface or using laser to measure without making contact. However, the existing technology is difficult to intuitively verify the shape and magnitude of the rail corrugation and sometimes the corrugation cannot be distinguished from defects on the rail surface. This study introduces a technology for detecting rail corrugation based on image analysis technique that is being developed to overcome the limitations of existing technologies. In addition, the analysis result is compared with the values measured from the contact sensor in order to verify the reliability of the image analysis technique applied to detect the rail corrugation. It is expected that by applying the rail corrugation detection technology based on the image analysis technique to the field, we can systematically build the information regarding the images, sizes, and shapes of the rail corrugations from the beginning of the development and utilize it to predict and prevent the occurrence of the corrugations as well as to develop corrugation reduction technology.