Toward online stability margin tracking of Turbomachines via operational modal analysis

• Stability margin estimation through Automated Operational Modal Analysis (AOMA). • Operational modal identification with vibration signals from a specific turbomachine . • Exploring AOMA for turbomachine condition monitoring. Proper design and monitoring of turbomachines are highly important for efficient and safe operation in industry. In the special case of turbomachines that contain seals, it is important to predict and avoid the steam whirl phenomenon. Steam whirl occurs when negative damping effects overtake positive damping effects, which leads to instability due to lateral vibration. The power level in which this phenomenon occurs sets the stability margin of the turbomachine. Predicting this phenomenon usually comprises modeling and testing machines during the design and commissioning stages. Design is quite challenging considering that accurate mathematical models are difficult to obtain and valid over a given operational range. Moreover, design and commissioning do not encompass changes due to deterioration and possible damage, which can change the stability margin during the machine's operation. Operational Modal Analysis (OMA) is used to extract modal parameters of systems during operation, with most applications related to civil structures. Turbomachines, in general, are non-linear and time-varying systems subjected to harmonic excitation, which hampers modal identification via OMA. Nevertheless, recent studies show that OMA's techniques have a lot of potential to support the condition monitoring of these systems. To contribute to this topic, this research proposes the estimation and tracking of the stability margin of a turbomachine with seals via Automated OMA (AOMA). Vibration signals from a turbomachine composed of a rotor supported by active magnetic bearings and containing annular gas seals are employed. The results show that different excitation conditions are required for modal identification and when adjusting these conditions, it is possible to extract good estimates of modes related to the foundation and the rotor, including the one that defines the instability margin. AOMA proves to be a promising tool for identifying the stability margin of turbomachines during operation, being valuable for condition monitoring.