Extraction of surface image features for wear detection on ball screw drive spindles
| Zugehörige Institution(en) am KIT | Institut für Produktionstechnik (WBK) |
| Publikationstyp | Proceedingsbeitrag |
| Publikationsmonat/-jahr | 11.2020 |
| Sprache | Englisch |
| Identifikator | ISBN: 978-3-7315-1053-6 KITopen-ID: 1000129215 |
| Erschienen in | Forum Bildverarbeitung 2020. Ed.: T. Längle ; M. Heizmann |
| Veranstaltung | Forum Bildverarbeitung (2020), Online, 26.11.2020 – 27.11.2020 |
| Verlag | KIT Scientific Publishing |
| Seiten | 305-316 |
| Bemerkung zur Veröffentlichung | Failures of production machines are often causedby wear and the resulting failure of components.There-fore, condition-based monitoring of machines and their com-ponents is becoming an increasingly important factor in in-dustry. Due to the simple conversion of the motion of elec-tric rotary drives into precision feed motion, the ball screwis an inherent element of many production machines. Thus,a failure of the ball screw often leads to costly productionstops. This paper shows the determination and extractionof wear-describing image features, allowing an image-basedcondition monitoring of ball screws using hyperparameter-optimized machine learning classifiers. The features to trainthe algorithms are derived and extracted based on the deepdomain knowledge of ball screw drive failures in combina-tion with further developed state of the art feature extractionalgorithms. |
| Schlagwörter | Ball screw drive, image features, artificial intelli-gence, machine learning, pattern recognition |
| Relationen in KITopen |