Analysis of Critical Conditions in Electric Power Systems by Feed Forward and Layer Recurrent Neural Networks

In this paper, critical conditions in electric power systems are monitored by applying various neural networks. In order to accomplish the stated goal, the authors tried several combinations of Feed Forward Neural Network and Layer Recurrent Neural Networks by imparting appropriate training schemes through supervised learning in order to formulate a comparative analysis on their performance. Once, training goes successful, the neural network learns how to deal with a set of newly presented data through validation and testing mechanism so as to evolve the best network structure and learning criteria. The proposed methodology has been tested on the standard IEEE 30-bus test system with the support of MATLAB based neural network toolbox. The results presented in this paper signify that the multi-layered feed forward neural network with Levenberg-Marquardt back propagation algorithm gives best training performance of all possible cases considered in this paper, thus validating the proposed methodology.