Domain Approximation for Condition Monitoring

We propose a novel algorithm for extracting samples from a data set supporting the extremal points in the set. Since the density of the data set is not taken into account, the method could enable adaptation to gradually changing data characteristics (e.g. machine wear). Using knowledge about the clustering structure of the data (obtained with multidimensional scaling techniques), the complexity of the solution can roughly be determined. 1 Introduction The condition of mechanical machinery can be monitored by measuring the vibration behaviour of its rotating parts [6]. Automatic recognition of machine wear and failure calls for methods that can deal with small sample sizes in highdimensional spaces, undersampled fault classes and dynamically changing environments. Since normal behaviour will probably be determined by a few calibration measurements of extremal operating conditions (e.g. when putting the machine into practice), an accurate but parsimonious description of the borders of ..