Volume 05,Issue 01

Online Distribution System's Voltage Stability Margin Monitoring Using Neural Networks and Optimization Algorithm

Authors

Nazmohammad Sarikhani, Arman Ghanbari Mazidi

Abstract
Due to the major blackouts caused by voltage collapse, the voltage stability problem has become one of the most significant challenges in the planning and operation of the modern electric power systems. Voltage instability has increasingly become a significant risk for the secure operation of electric power systems. In the heavily loaded electric power systems, the events causing voltage instability, lead to a progressive decrease in bus voltage magnitudes which may result in network islanding and blackout. For such systems, online voltage security assessment to preserve a desirable level of voltage security margin is a vital requirement for maintaining system security. This paper proposes an intelligent method for online voltage stability margin (VSM) estimation based on Radial Basis Function Neural Network (RBFNN) and association rules (AR). The proposed method includes four main modules: feature extraction, feature selection, estimator and training module. In the feature extraction module, the loading parameter of the network like voltage magnitude, active and reactive power have been extracted to be used as a raw input of estimator. In the feature selection module, the AR method has been employed to select the best set of the extracted features in the previous module. In the estimator module, RBFNN has been utilized and in the training module, bee's algorithm (BA) has been used to train the RBFNN. automatically select the number, location, and spread of basis functions to be used in RBF networks. The proposed method is applied to the New England 39-bus power system model and the obtained results have shown that the proposed method has excellent accuracy in VSM estimation.

Keyword: Voltage stability, RBFNN, Association rules, Estimation, Feature selection.

PDF [ 856.37 Kb ] | Endnote File | XML