Neural Network Based Assessment of Small-Signal Stability

Abstract

This paper presents a neural network (NN) based approach for the assessment of small-signal stability of single-machine infinite bus power system by predicting the modal components of the local mode. Two types of feedforward NNs were examined, namely, the radial basis function neural network (RBFNN) and back propagation neural network (BPNN). The input patterns of the generator real and reactive power, over a wide range of operating conditions, were associated with the corresponding output patterns of the damping factor and damped frequency of the local mode of oscillation. The performance of the NNs was evaluated in terms of the training and generalization simulation results. The performance comparison between the NNs shows that the RBFNN is much faster in training and more accurate in approximation than the BPNN. The simulation results also show that the proposed approach can be used as an effective tool by power system operators to predict the local mode and assess the small-signal stability of the system and take the appropriate action to enhance the system stability over a wide range of operating conditions by tuning power system stabilizers, etc.