General Information           Program          Registration

Claudio Canizares, University of Waterloo: Predictive Maintenance of Wind Generators Based on AI Techniques
Extended downtime of wind turbines is an issue that can be minimized by an optimal maintenance strategy via early fault detection. With large amounts of data collected through SCADA, machine learning techniques can be used to detect underlying failure patterns and notify customers of abnormal behaviour. A novel framework based on machine learning algorithms for fault prediction of wind farm generators will be presented for a wind farm in Summerside, PEI, evaluating the developed based on appropriate metrics. The results demonstrate the ability of the proposed techniques to predict wind generator failures, and their viability to optimize predictive maintenance strategies.

The Wind Energy Institute of Canada advances the development of wind energy across Canada through research, testing, innovation and collaboration.