Multi-Machine Equivalent Model for Impedance Analysis of Wind Farms

With the continuous expansion of wind farm scales and frequent occurrences of wide-band oscillations, establishing an accurate equivalent model that precisely captures the impedance characteristics of wind farms becomes imperative for investigating oscillation mechanisms and designing mitigation strategies. While existing equivalent methods primarily focus on maintaining time-domain consistency, they often fail to ensure frequency-domain consistency. To address this limitation, this paper adopts a multi-machine equivalent model that preserves the original wind farm topology. First, an improved K-means clustering algorithm integrated with the elbow method is employed to group turbines based on power grading principles. Subsequently, while retaining the inherent topological structure, the wind farm is equivalently transformed into a multi-group model comprising aggregated turbines paralleled with controlled current sources. Finally, frequency-sweep impedance simulations validate the accuracy of the adopted approach, demonstrating its superiority in impedance equivalence.