Abstract
AbstractOne of the most relevant renewable energy sources is wind power. The maintenance management strategy is necessary to ensure the reliability of wind farms. Supervisory control and data acquisition systems provide important status information of the wind turbine. The control monitoring systems generate a large amount of data, that requires extensive analytics, being useful to employ machine learning algorithms. A real case study is presented in this paper, using data from a real wind turbine to detect false alarms. The novelty proposed in this work is based on the implantation of tree classification models for the prediction and detection of false alarms. The results show an accuracy of 98.6%. The sensitivity and the specificity are 99.1% and 88.4%, respectively. These values indicate that the model used is effective for false alarm detection and identification.KeywordsWind turbine managementSCADAAlarmTree classification model
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call