Mise en place d’un estimateur d’endommagement de pale d’éolienne et de son espérance de vie résiduelle

Abstract

The thesis presented here contributes to the assessment of fatigue damage and the remaining life expectancy of wind turbine blades. This evaluation is based on the properties of the surrounding wind averaged over 10-min periods and recorded in the SCADA data of the studied wind turbine. To address the challenges of the energy transition imposed by a continuously growing energy demand and the gradual reduction of fossil fuels, energy providers are increasingly turning to wind power. With the continued growth of the global wind turbine fleet and the aging of existing wind turbines, operators are faced with the need to optimize maintenance planning to extend the lifespan of wind turbines while reducing associated costs. Among the critical components of wind turbines, blades are among the most susceptible to fatigue damage, underlining the importance of developing a model for estimating blade damage and remaining life expectancy. Firstly, we developed a wind turbine blade damage estimator that takes into account the average wind speed over 10-min periods, recorded in the SCADA data, as well as variations in wind speed at a frequency of 1 Hz. This high-frequency variability depends on turbulence intensity and average wind speed. Then we developed a tool to rapidly calculate blade damage for the entire lifespan of the wind turbine and compare relative damage between different wind turbines within the same wind farm. Secondly, our next contribution was to consider the different operating regimes of the wind turbine in damage evaluation. We demonstrated that damage caused by turbine stops and starts is negligible compared to the one caused when the turbine is in energy production mode over its entire lifespan. Finally, our last contribution was to adapt our damage estimator to assess the impact of different wind turbine operating profiles on its remaining life expectancy. We observed that reducing the wind turbine's nominal power can significantly extend this remaining life expectancy. This research was conducted in collaboration with Power Factors (Quebec, Canada). The results of this thesis are therefore based on data provided by Power Factors. Their involvement also enhances the potential interest of the industry in our project and its application to better estimate wind turbine blade damage due to fatigue.

Date	13 Sept 2024
Original language	French
Awarding Institution	École de technologie supérieure
Supervisor	Antoine Tahan (Supervisor) & Francis Pelletier (Co-supervisor)