Wind Farm Clustering Optimization Method Using Gap Statistic

Currently, multi-machine equivalent method is commonly used for large-scale wind farm simplify. Appropriate number of equivalent turbines not only ensures the equivalent accuracy but also reduces the complexity effectively. Therefore, it is necessary to optimize the number of equivalent turbines. A method to find the optimal number is proposed in this paper based on the Gap Statistic. The proposed method can directly evaluate the optimal clustering based on the operation data, and it is no longer necessary to compare the simulation results of equivalent models. And the factors that affect the accuracy of the equivalent model from a statistical and physical point of view is also discussed. A detailed model of test wind farm that combines 64 doubly fed induction wind generator turbines has been established based on DIgSILENT/PowerFactory. The simulation results show that the Gap Statistic method identified the "Elbow Point" of the change in equivalent precision successfully. Achieved the goal of balancing precision and complexity in wind farm equivalent.