PMU-ANN based real time monitoring of power system electromechanical oscillations

Abstract

Power system oscillations monitoring is a vital issue in operation of modern interconnected power systems. The existing methods for identifying the electromechanical modes are time-consuming and require modelling of the entire system that includes a large number of states and are performed offline. In this paper, an integrated Phasor Measurement Unit and Artificial Neural Network (PMU-ANN) based approach for online and real time monitoring of power system electromechanical oscillations is proposed. The placement of PMU is obtained using Integer Linear Programming (ILP). The data obtained from PMU is given as input to a multilayer Feedforward Neural Network (FFNN) and its output gives all the information related to the modes of the system and the mode ranking. The effectiveness of the proposed approach is investigated on IEEE 39-bus test system. The results show that the proposed approach is fast with less computational burden and is suitable for online and real time oscillations monitoring of the power systems under varying operating conditions.