Dynamic equivalent of wind farm model for power system stability studies

Abstract

A new dynamic equivalencing method for stability assessment of a grid-integrated wind farm is proposed in this article. The accuracy of the method is validated for a 34-bus system with 28-unit wind farm connected to Indian utility system. This wind farm consists of several wind turbines of two different ratings. The electrical parameters of the equivalent generator are derived from the mathematical model of the squirrel-cage induction generator. The parameters of the equivalent wind-turbine generator are optimized to yield minimum deviation from the detailed system response using genetic algorithm. The small-signal and transient stability responses of the study system with detail wind farm and equivalent model are simulated using MATLAB. Equivalent model eigenvalues are compared to the centre of inertia based detailed system eigenvalue. In addition, the computed eigenvalues and time-domain responses of the proposed equivalent model, detailed wind farm are compared against weighted model proposed earlier. In most of the investigated cases, the average error of dynamic responses between the proposed equivalent and detailed models are less when compared to weighted model. Thus, the large-signal responses of the proposed equivalent model show superior agreement with detailed system response.

Reduced-order Aggregate Dynamical Model for Wind Farms

“…Parametric scalings relating individual and aggregate models have been presented in the literature, see e.g., [4]- [6]. However, these works do not provide rigorous analysis to validate that the aggregate model synthesized with the scaled parameters faithfully captures the dynamics of the collection of individual turbines pointwise in time.…”