Input feature selection for real-time transient stability assessment for artificial neural network (ANN) using ANN sensitivity analysis

This paper presents a method for the selection of the input parameters, and their ranking for feedforward artificial neural networks (FF-ANN) applications in transient stability assessment. The method utilizes feedforward artificial neural networks to estimate the sensitivity of the output to all inputs. An evaluation of most of the common inputs used by the researchers is made. Sensitivity analysis using ANN is performed on key parameters to obtain the optimal ranking of the ANN input features. The critical clearing time (CCT) is used to assess the transient stability of the system. The proposed method is applied to a simple power system to illustrate the concept. The preliminary results show that the proposed sensitivity factors are converging to stable values.