Neural Network Based Approach for Steady-State Stability Assessment of Power Systems

The quest for an intelligence compliance system to solve power stability problems in real-time with high predictive accuracy, and efficiency has led the discovery of deep learning (DL) techniques. This paper investigates potency several artificial neural network (ANN) techniques assessing steady-state a system. new voltage pointer (NVSP) was employed parameterize reduce input data algorithms predict proximity systems instability. In this study, we consider five viz. feedforward (FFNN), cascade-forward (CFNN), layer recurrent (LRNN), linear (LLNN), Elman (ENN). evaluation is based on predictability accuracy these dynamic systems. were trained mimic NVSP dataset using Levenberg-Marquardt (LM) model. Similarly, performance analyses deduced from regression learner algorithm (RLA) root-mean-squared error (rmse) response plot graph. effectiveness NN demonstrated IEEE 30-bus Nigerian simulation results show that FFNN CFNN possess relatively better terms considered networks.