AJP Asia Intelligence Agency

Railway Research Institute Develops Automatic Rail Wear Diagnosis System for Moving Trains

A new technology has been developed to automatically check the wear condition of rails using a small inspection device installed on moving urban railway vehicles. Tests conducted on Incheon Urban Railway Line 2 at the operational speed of 80 km/h recorded a diagnostic performance exceeding 97%, raising expectations for its future use in reducing railway noise and vibration as well as in track maintenance.


On November 16, the Korea Railroad Research Institute (KRRI) announced the development of a 'vehicle-mounted rail image inspection device and diagnostic technology' capable of detecting rail corrugation occurring during urban railway operations.


Rail corrugation is a phenomenon where repetitive train operations create periodic irregularities on the rail surface. This wave-like wear can increase shocks and vibrations between the wheels and rails, leading to increased noise and decreased ride comfort.


In urban railways, this can be a source of noise complaints, particularly in areas adjacent to residential neighborhoods or in sharp curves. Therefore, it is crucial to continuously monitor the location and condition of wear and to perform maintenance, such as rail grinding, at appropriate times.


Previously, rail conditions were checked using expensive dedicated inspection vehicles or precision measuring equipment, which incurred high operational costs and faced limitations in conducting repetitive inspections in urban railway sections with many sharp curves and steep gradients.


To address this, KRRI developed a compact image inspection device utilizing a high-speed line scan camera. This device captures continuous high-resolution images of the 'rail head,' the top part of the rail where the wheels make contact, while the train is in operation.


When a vehicle equipped with the inspection device runs along the track, images of the rail surface are automatically collected, and a diagnostic program analyzes them to indicate the presence and location of rail corrugation, as well as representative wavelengths.


Operators can verify the actual rail images at the specified locations alongside the analysis results. Unlike previous methods that relied solely on numerical assessments of wear, this approach allows for direct observation of the rail surface condition to determine maintenance needs.


The inspection device has been developed in two configurations: an all-in-one PC type that operates independently without a separate control unit, and a separate PC type consisting of two devices weighing approximately 8 kg and 5 kg, respectively, for easier installation on railway vehicles.


KRRI collaborated with Incheon Transportation Corporation and Saeyoung Technology to equip electric trains and motor cars with the inspection devices and conducted field tests across the entire Incheon Urban Railway Line 2. The tests confirmed that high-resolution images of the rail surface could be obtained even at the maximum operational speed of 80 km/h.


The diagnostic performance for rail corrugation was reported to exceed 97% based on the comprehensive detection performance index known as the F1-score. This confirms the feasibility of applying the technology in actual urban railway operating environments, according to the researchers.


KRRI plans to further develop this technology by integrating image analysis with AI to provide comprehensive diagnostics for various defects, damages, and grinding conditions on rail surfaces, in addition to rail corrugation.


Hong Ji-young, a senior researcher at KRRI, stated, “With this technology, we can easily identify the occurrence and location of rail corrugation, which significantly impacts railway noise and maintenance.” She added, “We aim to enhance this into an intelligent rail condition management technology that supports predicting the occurrence and growth of rail corrugation and aids in maintenance decision-making.”


Sa Kong-myung, the head of KRRI, remarked, “This research achievement expands the field-centered technology for managing railway infrastructure conditions, which has been reliant on expensive dedicated inspection equipment, into a compact, lightweight, and continuous inspection method. We aim to develop practical technology that can simultaneously achieve urban railway noise reduction and improve rail maintenance efficiency, contributing to a quieter and safer railway environment that the public can experience.”





* This article has been translated by AI.