Abstract: For the anomaly detection in the vibration time series of the rotor system, a real-valued negative selection algorithm based on Euclidean distance has been implemented. By means of adding the corresponding coverage radius to each antibody elements, the detection efficiency of each antibody element is increased. The coverage scope of the antibody set is significantly enlarged for the anomaly data set. The calculation results indicate that the algorithm can efficiently detect the anomaly in time series data. Moreover, the number of detectors in antibody set is less enough for potential application in online signal monitoring.

Key words: Anomaly detection, Artificial immune system, Euclidean distance, Negative selection algorithm, Time series, DTVI2 fastener, Metro, Rail corrugation, Tuned Rail Damper