Impact of Auto-Correlation of Wind Speed on the Dynamics of Grid-Connected Doubly Fed Induction Generator

Abstract

This paper studies the impact of the auto-correlation of wind speed, which characterises the wind speed dynamics, on the electrical power system dynamics of a grid-connected DFIG system. Monte Carlo-based non-linear time domain simulations (MC-TDS) and stochastic eigenvalue analysis have been used to demonstrate the impact of stochastic wind dynamics. The statistical results of MC-TDS are validated using the stochastic calculus-based Lyapunov equation method (LEM). Moreover, the results are complemented by the Fourier analysis of corresponding TDS results. It is concluded that the autocorrelation of wind speed is an important factor characterising the influence of stochastic wind speed fluctuations on the dynamic performance of DFIG. The effect has been investigated with different values of grid strength, different drivetrain parameters of the DFIG and wind speed regimes. The performance of auxiliary damping controllers in suppressing the oscillations generated by stochastic variations in the wind speed has also been investigated.