Abstract
The classification of cyclically recorded time series plays an important role in measurement technologies. Example use cases range from gas sensors combined with temperature cycled operation to condition monitoring using vibration analysis. Before machine learning can be applied to high dimensional cyclical time series data dimensionality reduction has to be performed to avoid the classifier suffering from overfitting and the “curse of dimensionality”. This paper introduces a set of four complementary feature extraction methods and three feature selection algorithms that can be applied in a fully automatized manner to reduce the number of dimensions. The feature extraction algorithms are capable of extracting characteristic features from cyclical time series catching information contained in local details and overall cycle shape as well as in frequency or time-frequency domain. The methods for feature selection are capable of selecting the most suitable features for linear and nonlinear classification. The methods were chosen to be applicable to a wide range of applications which is verified by testing the set of methods on four different use cases.
Zusammenfassung
Die Klassifikation zyklischer Signalverläufe mittels maschinellen Lernens spielt eine wichtige Rolle in der Messtechnik. Beispielanwendungen reichen von Gas-Sensoren, die temperaturzyklisch betrieben werden, bis hin zur Zustandsüberwachung durch Vibrationsanalyse. Bevor maschinelles Lernen auf die hochdimensionalen, zyklischen Signalverläufe angewandt werden kann, muss deren Dimensionalität verringert werden, um zu verhindern, dass der Klassifikator unter Overfitting und dem ,,curse of dimensionality“ leidet. In dieser Veröffentlichung werden vier sich gegenseitig ergänzende Methoden zur Merkmalextraktion und drei Algorithmen zur Merkmalselektion vorgeschlagen, die automatisiert genutzt werden können, um die Dimensionalität zu verringern. Die Algorithmen zur Merkmalextraktion extrahieren charakteristische Merkmale aus dem Signalverlauf. Die in den Merkmalen enthaltene Information beinhaltet dabei nicht nur lokale Details und die allgemeine Kurvenform sondern auch Merkmale aus dem Frequenz- und Zeit-Frequenz-Bereich. Die Methoden zur Merkmalselektion sind in der Lage die besten Merkmale für lineare und radiale Klassifikation auszuwählen. Die Methoden wurden so ausgewählt, dass sie für ein möglichst breites Anwendungsspektrum geeignet sind, was durch die erfolgreiche Anwendung auf vier verschiedene Beispieldatensätze gezeigt wird.
About the authors
Tizian Schneider studied Microtechnologies and Nanostructures at Saarland University and received his Master of Science degree in January 2016. Since that time he has been working at Centre for Mechatronics and Automation Technology (ZeMA), division `Sensors and Actuators', in the field of data-based condition monitoring for industrial applications such as fluid power and electromechanical drive systems.
Saarland University, Dept. of Mechatronics Engineering, Lab for Measurement Technology, P.O. Box 15 11 50, 66041 Saarbruecken, Germany; and ZeMA – Center for Mechatronics and Automation Technology gGmbH, Eschberger Weg 46, Buisness Park, Building 9, 66121 Saarbrücken, Germany
Nikolai Helwig studied Mechatronics and received his diploma degree in 2013. Since that time he has been working at Centre for Mechatronics and Automation Technology (ZeMA), division `Sensors and Actuators', in the field of data-based condition monitoring for industrial applications such as fluid power and electromechanical drive systems.
ZeMA – Center for Mechatronics and Automation Technology gGmbH, Eschberger Weg 46, Buisness Park, Building 9, 66121 Saarbrücken, Germany
Andreas Schütze received his diploma in physics from RWTH Aachen in 1990 and his doctorate in Applied Physics from Justus-Liebig-Universität in Gießen in 1994 with a thesis on microsensors and sensor systems for the detection of reducing and oxidizing gases. From 1994 until 1998 he worked for VDI/VDE-IT, Teltow, Germany, mainly in the fields of microsystems technology. From 1998 until 2000 he was professor for Sensors and Microsystem Technology at the University of Applied Sciences in Krefeld, Germany. Since April 2000 he is professor for Measurement Technology in the Department of Mechatronics at Saarland University, Saarbrücken, Germany and head of the Laboratory for Measurement Technology (LMT). His research interests include microsensors and microsystems, especially intelligent gas sensor systems for security applications.
Lab for Measurement Technology, Saarland University, 66123 Saarbrücken, Germany, and Center for Mechatronics and Automation Technology (ZeMA), 66121 Saarbrücken, Germany
©2017 Walter de Gruyter Berlin/Boston
