Defect detection of railway pantograph contact strips based on laser scanning measurements

The safety of rail transport is of high importance to the economy. In particular, it refers to electric vehicles for which ensuring continuity of power supply is essential. The article presents a novel method for identifying defects in pantograph contact strips. The proposed method enables the detection of damage with a size comparable to the diameter of the contact wire, which could lead to a break to the catenary system and/or the pantograph itself. For this purpose, measurement data collected from the Laser Scanning Inspection System were utilized. The method is based on searching for local extrema in derivatives representing abrupt thickness changes of a contact strip, e.g. those caused by material chipping. This process is carried out using the Savitzky-Golay filter and an algorithm for peak detection. Moreover, the implementation of the proposed method eliminates the need for human intervention in diagnosing contact strip damage. It facilitates the implementation of predictive maintenance, replacing the commonly used preventive and time-based maintenance strategies. The developed method was verified and validated through computer simulations in Matlab, which allowed for the testing and parameterization of computational modules. Subsequently, a dataset collected under laboratory conditions, containing artificially introduced damage of known dimensions, was analysed. Finally, scanning results from a measuring system installed on a railway line were examined. The representative cases were selected: a new, damaged, and destroyed contact strip. The presented results demonstrated the effectiveness of the described method.