Abstract:

This study proposes a parameter estimation method, together with a grid synchronisation algorithm, for wind turbine-driven doubly fed induction generators (DFIGs). Aiming at achieving an automated control procedure, their integration with a previously published power control strategy is also addressed. During a normal operation mode, the DFIG is grid-connected and generated power is commanded by combining a sliding-mode control (SMC) scheme, which provides high dynamic performance and robust behaviour, with a model reference adaptive system observer, estimating both rotor position and speed without the use of mechanical sensors. In order to preserve performance during start-up and grid connection without the additional requirement of an encoder, the proposed parameter estimation and grid synchronisation schemes are both sensorless. Moreover, the same SMC structure of the power control strategy is also adopted for the grid synchronisation algorithm, which facilitates transfer between synchronisation and power controllers at the instant of grid connection. Thereby, a global sensorless SMC configuration result, which is self-matched by the parameter estimation process. The resulting scheme has been applied to a hardware-in-the-loop-based DFIG virtual prototype under realistic wind conditions, obtaining satisfactory results.