Artificial Neural Network-Based Poles Estimation for Online Stability Analysis of Wind Farms

Online small-signal stability analysis of wind farms is the key to activate preventive oscillation suppression measures. This paper proposes a near real-time poles estimation method for the online stability margin calculation via the Artificial Neural Network (ANN). First, based on the nodal admittance transfer function matrix of the wind farm, a numeric poles calculation algorithm is employed to efficiently generate the training set for the ANN. Then, a mode shape cluster-based poles filter method is proposed to improve the quality of the training set. At last, the ANN model is designed and trained with input as each turbine's active/reactive power, voltage phasor and ac grid short circuit ratio (SCR), while output as the dominant pole. The effectiveness of the ANN-based poles model is validated on a 100MW wind farm by comparing with real-time simulation results.