Static Security Assessment of Power System Using Radial Basis Function Neural Network Module

Abstract

Contingency screening and ranking method has gained importance in modern power system and plays a crucial role in its secure operation. This paper proposes Radial Basis Function Network (RBFN) to realize static security assessment of power system using regression module. Newton-Raphson Load Flow (NRLF) is used for load flow analysis and single line outages are considered for pattern generation. Composite Security Index (CSI) is calculated to assess the severity of the system, which considers the combined effect of the limit violations of bus voltages as well as power flow of the transmission lines. After training, module predicts the Composite Security Index and ranks those in descending order based on Composite Security Index. Less number of variables are selected from total variables by Single Ranking and Correlation Coefficient method. The proposed module is applied to IEEE 14, IEEE 30-bus and IEEE 57-bus standard test system, where the testing results prove its performance and robustness for static security assessment of power system. The comparison of ranking obtained by Radial Basis Function Network and the Newton-Raphson Load Flow analysis in terms of time and accuracy is presented. It is found that the proposed model is quick and accurate to predict static security assessment of power system.