Bulb turbine anomaly detection using machine learning

The current global energy landscape is characterized by soaring demand. To satisfy this, hydroelectric plants face the challenge of operating under intense conditions that can present challenges for effective operation and maintenance. Consequently, the imperative to minimize downtime and swiftly identify faults is an ongoing priority. The occurrence of failures can lead to costly downtime and result in significant financial, material, and even human losses for the operating company. It is therefore of significant interest to explore opportunities that may enable the prevention of such failures, thereby shifting from traditional corrective and preventive maintenance to a more proactive approach of predictive maintenance. In this paper, an unsupervised machine-learning approach is proposed, that aims to verify if an isolation forest model method can automatically extract useful information from signals and detect anomalies in the bulb-type turbine regulation ring.