AUTOMATICALLY GENERATING TRAINING DATA OF A TIME SERIES OF SENSOR DATA

Assistance device (300) for automatically generating training data of a time series of sensor data, further on called temporal sensor data, applied to train an Artificial Intelligence system used for detecting anomalous behaviour of a technical system (100), comprising at least one processor configured to perform- obtaining (S1) historical temporal sensor data measured at said or a similar technical system (100),- dividing (S2) the historical temporal sensor data into a temporal sequence of segments and assigning one segment type out of several different segment types to each segment, wherein each segment type is characterized by similar data distribution of temporal sensor data,- iteratively (S3) for each segment, determining a neighbourhood pattern of segment types comprising of the segment type of a first number of adjacent preceding segments and the segment type of said segment and the segment types of a second number of adjacent subsequent segments,- determining (S4) the most frequently occurring neighbourhood pattern from all determined neighbourhood patterns as reference pattern for normal operation of the technical system,- selecting (S5) a subsequence of segments out of the historical temporal sensor data, which is ordered according to the reference pattern, and- outputting (S6) the subsequence of segments for applying as training data.