Abstract This research paper introduces advancements in hydraulic turbine condition monitoring through the implementation of vibration-based condition indicators. Conventional methods rely on vibration monitoring systems, triggering alarms or unit trips based on overall measured values at specific points. However, these methods may generate false alarms under changing operating conditions, such as variations in head or flow rate. In response, our study focuses on the development and validation of condition indicators derived from vibration data, considering the turbine’s operating conditions. Our methodology defines and validates these condition indicators, providing a detailed understanding of the hydraulic turbine’s condition. By employing an enhanced condition monitoring framework, our approach not only improves fault detection accuracy but also offers a comprehensive perspective on the overall health of the turbine. Experimental results from a prototype Kaplan turbine showcase the effectiveness of our methodology in detecting and characterizing faults, enabling proactive maintenance strategies. The Kaplan turbine studied is one of the demonstrators of the XFLEX HYDRO project. In conclusion, this research delivers a valuable tool for industry professionals and researchers involved in hydraulic turbine maintenance and optimization.
Read full abstract