Conception et comparaison des performances de strategies de commande PI et DAC appliquees au calage variable d'une eolienne de 10kw

One of the goals of the Wind Energy Strategic Network (WESNet) is to optimize wind turbine technologies for the Canadian climate. The 10kW wind turbine was developed at the New-Brunswick University and the Ecole de techonologie superieure to fulfill this mandate. It is equipped with an electromechanical variable pitch system and a generation system using torque control to maximize its power output. Unlike what's usually seen in the industry, this last control method remains active for wind speeds above the nominal value.;This project aims to design two control strategies for the variable pitch system of the WESNet wind turbine and to compare their performances under operating conditions in order to identify the one that yields the higher power output. They are compared through the means of their power curve, their availabilities and their realistic extrapolated annual energy production. This last indicator is obtained by multipling one strategy's extrapolated annual energy production by its availability. It gives an overview of a strategy's potential production.;The control strategies are developed with a linear and a nonlinear model of the wind turbine built with SIMULINK and the FAST wind turbine design code. They are then implemented in the variable pitch drive and data of their operation is recorded during field test lasting for approximately a month in each case.;The first strategy developed is proportional-integral (PI). It is chosen because of its simplicity of design and application, the amount of available literature and of its proven performances. Field tests as shown that a gain scheduling technique was required in order to obtain good power regulation. The data recorded was used to validate the nonlinear model behavior.;The second strategy is disturbance accommodating control (DAC). It is chosen because of its flexibility, of its relative simplicity for a modern control strategy and of the promising experimental results found in the literature.;The wind turbine produces more when equipped with the DAC strategy then the PI strategy. The DAC power curve for winds above the nominal speed is higher and more constant : the 10kW production was only reached with the DAC. Furthermore, despite a lower availability, it exhibits a higher realistic extrapolated annual energy production than the PI. This improved production is caused by an increased pitch activity. The DAC strategy exhibits a better regulation for the majority of wind conditions, while PI is better when winds are turbulent (above 18% IT). However, the PI fails to produce more than the DAC for all wind conditions.