Analyse des sous-performances énergétiques d'éoliennes en utilisant des cartes de contrôle EWMA

Abstract

Wind energy is growing rapidly for over more than two decades, many wind farms were built during this period. Nowadays, wind farms have to be monitored in order to obtain the best performance as possible, ensuring optimal operating conditions and high availability. As technological advances of past few years were not existing when wind farms were built, tools have to be developed in order to detect underperformance of wind turbines as soon as possible, and correct problems in a prevent way. In a first part, this thesis details the steps to build wind turbines power curves when machines are operating normally. From data collection, coming directly from wind turbines, to final power curve which is used to detect underperformances effectively, all the process is explain step by step. In a second part, an exponentially weight moving average control chart was developed. Thanks to a graphical method and an algorithm, it is able to detect small shift on the power of the wind turbine as time goes by. The control chart enables to detect gradual shift of about 1 % over one year, compare to the normal behavior of the wind turbine.

Date	2 Jul 2014
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Christian Masson (Supervisor) & Antoine Tahan (Co-supervisor)