Railway switches and crossings (S&Cs) are among the most important high-value components in a railway network and a single failure of such an asset could result in severe network disturbance, huge economical loss, and even severe accidents. Therefore, potential defects need to be detected at an early stage and the status of the S&C must be monitored to prevent such consequences. One type of defect that can occur is called a squat. A squat is a local defect like a dent or an open pit in the rail surface. In this thesis, a testbed including a full-scale S&C and a bogie wagon was studied. Vibrations were measured for different squat sizes by an accelerometer mounted at the point machine, while a boggy was travelling along the S&C. A method of processing the vibration data and the speed data is proposed to investigate the feasibility of detecting and quantifying the severity of a squat. A group of features were extracted to apply isolation forest to generate anomaly scores to estimate the health status of the S&C. One key technology applied is wavelet denoising. The study shows that it is possible to monitor the development of the squat size introduced in the test bed by measuring point machine vibrations. The relationships between the normalised peak-to-peak amplitude of the vibration signal and the squat depth were estimated. The results also show that the proposed method is effective and can produce anomaly scores that can indicate the general health status of an S&C regarding squat defects.
|Publisher||Luleå tekniska universitet, Drift, underhåll och akustik|
|Source Sets||DiVA Archive at Upsalla University|
|Type||Licentiate thesis, comprehensive summary, info:eu-repo/semantics/masterThesis, text|
|Relation||Licentiate thesis / Luleå University of Technology, 1402-1757|
Page generated in 0.0029 seconds