부산대학교 도서관

Although numerous valuable research has been conducted by many authors, the relation between evolving mechanical faults and their fault-indicating reflections in the electrical signals of rotating electric machinery still requires extensive attention. This lack of information implies serious obstructions in the evolution of applying motor current signature analysis as a complete and reliable condition monitoring technology. However, as previous research of the authors resulted in the construction of a unique test-bench where specific mechanical faults can be introduced into an 11kW induction machine with high reproducibility and accuracy, the relations between evolving mechanical faults and their reflections in the stator current can be investigated and quantified experimentally. Using this mechanical fault-emulator, experiments where conducted where the severity of a single point outer race bearing fault and its corresponding signature in the stator current's spectral fault-components are accurately analyzed. These evolving faulty components are trended and characterized, resulting in the desired quantification of the mechanical fault-severity. Based on these results, it is possible to inversely estimate the severity of single point outer race bearing problems by acquiring nothing more then the stator current time-functions.