A dynamic equivalent method for DFIG wind farm based on the cloud model of rotor current

Abstract

Affected by low voltage ride-through(LVRT) characteristics, random wind speed and terminal voltage dips, large scale doubly-fed induction generator (DFIG) wind farm has a dynamic characteristic of randomness and fuzziness. According to the theory of uncertainty mathematics, this paper presents a dynamic equivalent method for wind farm based on rotor current cloud model. The peak value of rotor current of DFIG is chosen as cloud droplet sample, which can be obtained by simulation under random wind speed and terminal voltage dip. The wind turbine generators can be clustered by three feature indices through backward cloud generator, including expectations, entropy and hyper entropy. Finally, the simulation based on DIgSILENT/PowerFactory is carried out. The results show that the proposed equivalent model can accurately reflect the transient fault characteristics of the detailed wind farm model, which is suitable for analysing and calculating the short-circuit current of wind farm.