From 4th to 6th September, our colleague David López presented his final degree project on smart control entitled “Automatic system for detecting failure alarms in the mechanical gear of a wind turbine” at the automation conference “XL Jornadas de Automática 2019”, an event organised by the Comité Español de Automática (CEA).

This work presented an automatic system for detecting alarms in the mechanical gear of a wind turbine in a wind farm. This system is based on a digital twin, in other words, a simulated physical model of the case study (in this case, a transmission system), to practise with simulated system data. This is accompanied by a trained detection model based on classification techniques which, using real data, returns the device’s failure alarms. Hence, cover in the event of failures is maximised, minimising potential positive and negative false readings in the system.  

This work includes the study of failures in a bearing, with a procedure which can be applied to a mechanical gear and in turn to a wind turbine, thus making it possible to significantly optimise the predictive detection of failures as well as wind turbine maintenance tasks.

Congratulations David!